Datasets:

sulovexin

/

mysummary

like 0

journal.ppat.1006509

CD4 is expressed on a heterogeneous subset of hematopoietic progenitors, which persistently harbor CXCR4 and CCR5-tropic HIV proviral genomes in vivo

Long term combination anti-retroviral therapy ( cART ) blocks viral spread in vivo but is not curative , as plasma virus rebounds after cART interruption ., Sequence analysis of residual circulating and rebounding virus in HIV+ patients indicates that virions likely come from the activation of latent provirus that had been archived since before the initiation of therapy rather than from low-level replication and spread of cART-resistant virus 1 , 2 ., HIV enters cells via HIV Env interacting with CD4 plus a co-receptor , usually CCR5 or CXCR4 ., CXCR4-utilizing viruses differ from those that utilize CCR5 in their ability to infect stem cells that can engraft and generate multiple lineages in a mouse xenograft model 3 ., In contrast , CCR5-tropic viruses infect HSPCs that are restricted in their capacity to differentiate 3 ., Recently , Nixon and colleagues elegantly demonstrated that myeloid progenitors , including common myeloid progenitors ( CMPs ) and granulocyte/monocyte progenitors ( GMPs ) , express CCR5 and can be infected by CCR5-tropic HIV in vitro and in a humanized mouse model 4 ., Based largely on patterns of hematopoiesis that occur following transplantation , hematopoietic progenitors , such as those targeted by CCR5-tropic HIVs , were thought to be short-lived in vivo 3–5 ., However , in situ tagging experiments in mice have recently found that non-stem cell progenitors make an enduring contribution to native hematopoiesis in adults through successive recruitment of thousands of clones , each with a minimal contribution to mature progeny 6–8 ., Consistent with this , non-stem cell myeloid progenitors such as GMPs were found to persist in people with aplastic anemia despite dramatic losses of stem cells 6 ., Thus , a large number of long-lived progenitors , rather than classically defined Hematopoietic stem cells ( HSCs ) , may be the main drivers of steady-state hematopoiesis during adulthood 7 , 8 ., Here , we provide evidence that non-stem cell hematopoietic progenitors harbor CCR5-tropic HIVs for years in optimally treated people , providing new evidence that non-stem cell progenitors are long-lived in people without evidence of bone marrow disease and can potentially serve as reservoirs of HIV ., We also demonstrate that CD4high HSPC subsets that we show include multi-potent progenitors ( MPPs ) are preferentially targeted by both HIV subtypes in vitro ., Moreover , we provide in vivo evidence that infected HSPCs can differentiate into multiple lineages that harbor provirus ., These data expand our understanding of HIV infection and hematopoiesis by demonstrating that in addition to stem cells , intermediate progenitor cells potentially provide an enduring reservoir for CCR5- and CXCR4-tropic HIV proviral genomes ., To better understand the types of hematopoietic stem and progenitor cells ( HSPCs ) that are infected by HIV in vivo , we developed an approach to efficiently isolate HSPC populations enriched ( Sort, 1 ) or depleted ( Sort, 2 ) for stem cells ( Fig 1A and 1B ) ., Compared to Sort 1 cells , Sort 2 cells expressed lower levels of CD133 ( Fig 1B ) and were depleted for hematopoietic stem cells ( HSCs ) and multi-potent progenitors ( HSC/MPPs ) ( Fig 1C–1G ) ., Conversely , Sort 2 cells were enriched for more restricted progenitors ( common myeloid progenitors ( CMPs ) and megakaryocyte/erythrocyte progenitors ( MEPs ) ( Fig 1H and 1I ) 9 ., Enrichment of MEPs in Sort 2 samples was confirmed using methylcellulose colony formation assays ( Fig 1J ) ., To develop a better understanding of which HSPCs harbor HIV in vivo , we obtained samples from 47 HIV-infected donors , including two that had been initially treated during acute infection ., All donors were on therapy with undetectable viral loads for least six months ., A 20 ml bone marrow sample and 100 ml of peripheral blood were collected from each donor ., HSPCs were isolated from adherence depleted bone marrow mononuclear cells in two steps as described in Fig, 1 . From 20 cc of bone marrow , we obtained ~2 . 5x106 total HSPCs per donor ., For 41 of 47 donors , we obtained adequate aspirates and the purified HSPCs met our criteria of having <1% CD3+ T cell contamination and >80% CD34+ or CD133+ cells ., The mean purity of included samples was approximately 94% CD133 for Sort 1 and 90% CD34 for Sort 2 ( Table 1 , S1 and S2 Tables ) ., DNA was isolated from each sample and multiplex single genome amplification ( SGA ) polymerase chain reaction ( PCR ) was used to amplify gag and env amplicons or near full-length genomes ., For each donor , we selected a primer pair combination that most efficiently amplified HIV sequences from peripheral blood mononuclear cell ( PBMC ) DNA prior to testing HSPC samples ., After analyzing at least 80 , 000 cells from all samples that met our purity criteria , we determined that most donors ( n = 24 , 59% ) had detectable HIV provirus in HSPCs ., More cells were screened in the positive group than in the undetectable group ( 975 , 959 versus 661 , 965 ) but that difference and the level of sample purity between the two groups were not statistically significant ( Table 1 ) ., Further , the timing of HAART was not a significant factor in our ability to detect provirus; one of two donors treated since acute infection had detectable HSPC-associated provirus and provirus was present in long term suppressed patients ( up to 9 . 8 years , Table 1 ) ., The overall mean frequency of provirus in HSPCs was 2 . 4 copies per million cells based on the number of positive 1st round PCR reactions ( 82 ) set up at limiting dilution that produced a gag and/or env amplicon out of the total number of cells assayed ( 35 million ) ., For individual donors , the frequency ranged from <1 per 1 . 3 x106 cells to 18 copies per 106 cells ., To rule out T cells as a source of HIV DNA in HSPC samples , we eliminated all HSPC samples with >1% contaminating CD3+ cells and all samples included in our final analysis contained <0 . 52% CD3+ cells ( S1 and S2 Tables ) ., In addition , we used a previously published statistical method that takes into account HIV genome frequency in sorted and flow-through samples , assigning a p value to indicate the likelihood that HIV DNA in HSPC samples came from T cells 10 ., This analysis is shown in detail in Fig, 2 . Briefly , we carefully assessed the frequency of CD3+ T cells and provirus in both the sorted sample and in the flow-through sample ., Then , we compared the frequencies assuming that only CD3+ T cells account for all provirus ., As shown in Fig 2B , the frequency of infected CD3+ T cells would have to have been much higher in the sorted sample than in the flow-through to account for the provirus in the sorted HSPC samples ( e . g . 1 in 52 versus 1 in 15 , 000 for donor 409000 ) ., This difference is assigned a p value that takes into account 95% confidence intervals and only samples with p<0 . 05 were included in our final analysis ., Consistent with our conclusion that HIV DNA from Sort 1 and 2 came from HSPCs and not CD3+ T cells , we observed no correlation between proviral frequency in the samples and the frequency of contaminating CD3+ cells ( S2 Fig ) ., In addition , rearranged T cell receptor PCR assays were performed to confirm that near-full-length genomes from HSPC DNA samples were unlikely to have originated from T cells ( Fig 2C ) ., Similar results were obtained from donor 413402 , which was screened by PCR because too few cells were available to accurately assess this sample by flow cytometry ( S4 Table , S3 Fig ) ., The caveats for the rearranged TCR PCR assay are that it is not quantitative and it is associated with non-specific background bands that limit the amount of DNA that can be added to the reaction . These non-specific bands arise in all samples , including negative control HEK 293 cells , and are not related to TCR based on sequencing analysis . Because of these limitations , the statistical analysis we described in Fig 2 provides a more robust and quantitative assessment . Based on initial results that CCR5-tropic HIVs infect non-stem-cell progenitors that were originally believed to be short lived , we expected to mainly observe CXCR4-tropic virus in HSPC preparations ., To assess this , we examined env amplicons available to study from a subset of 19 donors from the overall cohort ., As summarized in Tables 2 and 3 , we isolated a total of 52 env C2-V3 amplicons ., Each amplicon was assigned a genotype using the indicated co-receptor prediction software ( Table 3 ) ., 16 amplicons from 8 donors were predicted to be CXCR4-tropic , including three near-full-length genomes with full open reading frames and cis elements ( S3 Table ) ., Unexpectedly , we also isolated a total of 36 amplicons from 17 donors that were predicted to be CCR5-tropic , including one near-full-length genome with full open reading frames and cis elements ( S3 Table ) ., Overall , the genotopyes of env amplicons from HSPCs closely matched those from peripheral blood mononuclear cells for each donor ( Table 3 ) ., Because env genotype prediction tools are not always reliable , we confirmed Env tropism with a phenotypic assay ., For this analysis , we used either HSPC-derived full-length Env or a non-HSPC-derived Env with identical nucleotide or amino acid V3 region from the same donor as available ( Table 4 ) ., A phenotypic assay utilizing 3T3 cells expressing CD4 and individual chemokine receptors 11 was used for this assessment ., This assay confirmed the tropism of ten CCR5-tropic Envs , four CXCR4/dual tropic Envs and demonstrated that one Env was not functional ., The isolation of HIV encoding Envs of both tropisms from HSPCs suggests either that CCR5-tropic Envs unexpectedly target HSCs or that restricted progenitor cells targeted by CCR5-tropic viruses survive longer in vivo than expected ., To better understand whether CCR5-tropic viruses might target restricted progenitors that persist longer than expected , we asked whether provirus could be detected in Sort 2 , which contained restricted progenitors that were unlikely to be stem cells ., Interestingly , we found no significant difference in the number of donors with amplicons in Sort 1 versus Sort 2 subsets 14 donors had amplicons isolated from Sort 1 and 11 had amplicons isolated from Sort 2 , ( Tables 2 and 3 ) ., The mean frequency of env amplicons was higher in Sort 1 than Sort 2 but this difference did not achieve significance ., ( The mean frequency was four copies per million cells for Sort 1 versus two copies per million cells for Sort 2 , p = 0 . 06 . ), None of the amplicons isolated from Sort 2 were identical to those from Sort 1 , indicating that independent infection of restricted progenitors rather than differentiation of infected stem cells explains the presence of provirus in this population ., In sum , these results suggest that non-stem cell restricted HSPCs can be infected by HIV and endure for at least the period of effective antiretroviral treatment ., The result that HSPCs depleted of HSCs harbor HIV provirus that persists in optimally treated people as well as the finding that CCR5-utilizing virus persists in HSPCs was unexpected; therefore , we pursued additional evidence to better understand this finding ., First , we assessed expression of HIV receptors in HSPC subsets ., To accomplish this , we used a publicly available microarray dataset of RNA expression in human bone marrow HSPCs 12 and used established markers to purify murine bone marrow HSPCs for an RNA-seq analysis to profile expression of HIV receptors in HSPC subtypes ., After confirming that progenitor subsets from each species expressed the expected developmentally appropriate set of genes ( Fig 3 ) we found that both approaches yielded similar results ., As shown in Fig 3 , both revealed very low CCR5 expression in HSCs with higher expression in some restricted hematopoietic progenitor sub-types ., These results agree with published studies showing low or no expression of CCR5 protein by HSC-enriched cells 3 , 13 with more expression of CCR5 protein in restricted hematopoietic progenitor sub-populations 4 , 13 ., In addition , both approaches showed that CXCR4 and CD4 RNA were expressed by HSCs and several other progenitor populations ( Fig 3 ) ., Based on this analysis , CXCR4-tropic viruses are predicted to target a wide range of progenitor subsets including HSCs whereas CCR5-tropic Envs are more likely to target restricted HSPC subsets such as GMPs ., In prior studies , we used pseudotyped lentiviral reporter constructs to examine differential targeting of HSPCs by CCR5 and CXCR4 and it remained possible that full length , wild type HIVs target cells differently ., To examine this question , we compared HIV infection of HSPCs by two wild type viruses , NL4-3 ( CXCR4-tropic ) and YU-2 ( CCR5-tropic ) ( Fig 4A ) ., After demonstrating that CD133bright cell populations contain the majority of HSCs based on CD38 , CD45RA and CD90 staining ( Fig 4B ) , we used the level of CD133 staining to assess HIV infection of HSCs ., As shown in Fig 4C , full length HIVs demonstrated the same pattern as previously observed using pseudotyped lentiviral vectors; CCR5-tropic YU2 infected a restricted pattern of progenitors depleted of stem cells whereas NL4-3 targeted a wide range of progenitors , including those likely to be stem cells ., ( Maraviroc and AMD3100 appropriately inhibited entry via CCR5 and CXCR4 respectively , Fig 4C , lower panels . ), Correspondingly , on average , we measured about 4 . 5 times more CD133 on HSPCs infected by NL4-3 than those infected with YU2 ( Fig 4D and 4E ) ., Further , the same pattern was observed using a lentiviral construct ( HIV-7/SF-GFP ) pseudotyped with additional Env proteins including one from a CCR5-tropic transmitted/founder virus SVPB16 ( SV16 ) 3 , 14 , 15 ( Fig 5 ) . In sum , consistent with prior results , CCR5-tropic viruses consistently demonstrated a restricted pattern of infection of more differentiated progenitors that contrasts with the wide range of progenitors targeted by CXCR4-tropic and VSV-G-pseudotyped viruses . Confirmation that a wide range of CCR5-tropic HIVs are restricted to non-stem cell HSPCs suggests that the CCR5-tropic HIV we detected in stem cell–depleted HSPC populations from patients likely came from more restricted progenitors that survived longer than previously appreciated and that these cells might also serve as long lived cellular reservoirs of HIV . The HIV receptor , CD4 , is usually required for infection and is expressed on CD34+ HSPCs , although at low levels compared to CD4+ T cells 16 , 17 ., If the relative level of CD4 expression on HSPCs determined susceptibility of HSPCs to infection , then CD4 expression would serve as an indicator of the subtypes of HSPCs potentially targeted by HIV ., To examine this question , we treated HSPCs with a GFP-expressing lentiviral vector pseudotyped with CCR5- or CXCR4-tropic Env proteins ( Fig 6A and 6B ) and assessed CD4 levels on the GFP+ transduced cells ., We observed that HSPCs within a CD4high flow cytometric gate displayed 2–30 times greater infection than CD4low/- cells ( Fig 6A–6C ) ., The increased infection of CD4high cells was not due to a greater capacity of these cells to support infection by this virus because the same virus pseudotyped with the vesicular stomatitis virus glycoprotein ( VSV-G ) demonstrated no such preference ( Fig 6B ) ., Further , CCR5-tropic envelopes had a significantly greater propensity to target CD4high progenitors compared to CXCR4 and dual-tropic envelopes ( Fig 6B and 6C ) ., Thus , relative CD4 expression levels correlated with susceptibility of HSPCs to infection by HIV-1 and HSPCs that express higher levels of CD4 are more likely to become infected ., To determine whether CD4 marks a stable and separable progenitor subset with unique characteristics , we used flow cytometry to determine whether HSPCs could be separated into low and high CD4 expressing cells ., Remarkably , sorting separated two distinct HSPC populations with different levels of CD4 ( Fig 7A ) ., We then used these populations to demonstrate that CD4high HSPCs could form GEMM , granulocyte/macrophage ( GM ) , and erythroid ( E ) colonies ( Fig 7B ) ., Thus , CD4 marks a subset of HSPCs that includes a number of different types of progenitors , including multipotent progenitors capable of generating multi-lineage GEMM colonies ., To examine the CD4high sub-population in more detail , we used cell surface markers that had been validated with functional assays for HSPC subsets 9 ., Remarkably , we found that CD4high HSPCs in Sort 1 contained a significantly greater frequency of HSCs and MPPs ( CD38-CD10-CD45RA- ) than CD4low HSPCs in the same Sort ( Fig 8 ) ., Because CD133 also marks populations enriched for HSCs , we confirmed this result by demonstrating that there were significantly higher levels of CD4 on CD133high HSPCs than on CD133dim HSPCs ( ratio paired t test , p = 0 . 020 ) ., In contrast , Sort 1 CD4low/- HSPCs and all Sort 2 cells that had lower levels of CD133 ( including those that were relatively CD4high ) were less frequently HSC/MPPs and more frequently restricted progenitors such as CMP/MEPs ( Fig 8 ) ., Similar results were obtained whether or not lineage positive cells were depleted from the sample prior to analysis ( Fig 8C , open symbols ) ., Thus , CD4 is expressed by a heterogeneous subset of hematopoietic progenitors and is expressed at significantly higher levels on subsets that include HSCs and MPPs ., If HIV infects progenitor cells in vivo , HIV genomes could theoretically be passed to differentiated daughter cells as long as differentiation did not lead to reactivation of the virus from latency and cell death ., To determine whether HIV can be transmitted by differentiation of infected progenitors , we assessed HIV proviral frequency in CD4-negative HSPC progeny ., ( CD4-negative progeny were chosen for this analysis because cells lacking this HIV receptor are unlikely to be directly infected . ), To reduce the possibility of contamination by CD4-expressing cells , we depleted CD4+ cells using an anti-CD4 magnetic bead column prior to fluorescence activated cell sorting ( FACS ) ., Following bead depletion and FACS , CD3+CD4+ T cells were undetectable in most samples ( Table 5 ) ., Moreover , lineage-positive cells ( CD19+ B cells , CD8+ T cells and CD56+ natural killer ( NK ) cells ) were >98% CD4 negative ( indicated as “post-FACS” in Fig 9A ) ., To determine whether HIV proviral DNA was present in these lineages , we used multiplex SGA PCR as described above ., Remarkably , we generated a total of 38 LTR-gag or C2-V3env amplicons from four of five donors with CXCR4-tropic HIV but only one of five donors with only CCR5-tropic virus ( Table 5 ) ., In two cases ( donors 420000 and 431000 ) , amplicons were identical to those isolated from HSPCs ( indicated as # in Table 5 ) ., These cells were highly purified with undetectable CD3+CD4+ T cells ( Fig 9A and Table 5 ) ., Using a quantitative statistical analysis , we found that the amplicons from CD4-negative lineages were unlikely to have come from contaminating CD3+CD4+ T cells ( p<0 . 05-p<0 . 001 , Table 5 ) ., These results provide , strong evidence that HIV provirus can be transmitted from infected progenitors to progeny cells in vivo ., Although we only detected provirus in CD4-negative cells from one of five donors ( 431000 ) with predominantly CCR5-tropic HIV , this donor provided the strongest evidence for HIV infection of multi-potent progenitors ., Indeed , using SGA PCR , we amplified 14 identical CCR5-tropic C2-V3env amplicons from all three CD4-negative lineages , which were perfect matches to one another as well as to an amplicon isolated from HSPCs Figs 9B and 10A ., In addition , seven first round SGA multiplex PCR reactions generated both C2-V3env amplicons as well as LTR-gag amplicons , all of which were identical ( Table 5 , Figs 9C and 10A ) ., Remarkably , these amplicons contained a signature 469 bp deletion that removed the packaging site , the major splice site and the gag start codon , effectively inactivating the virus ( Fig 10A ) ., We confirmed that the deleted gag came from the same proviral genome as the C2-V3env amplicons by using SGA PCR to isolate two near-full-length genome amplicons from CD4-negative cells ( Fig 10A ) ., The presence of replication defective clonal proviral genomes in multiple differentiated hematopoietic lineages and in HSPCs provides strong evidence that infected multi-potent progenitors persist and differentiate in optimally treated people ., A phylogenetic analysis of all donor sequences ensured that all donor 431 sequences clustered together , ruling out contamination and cross contamination as confounding factors ( S1 Fig ) ., Moreover , phylogenetic analysis revealed that amplicons isolated from CD4-negative cells ( B , NK and CD8 ) were not common in CD4+ cells or unfractionated PBMCs , making cross-contamination an unlikely explanation for their relatively high frequency in CD4-negative lineages ( Fig 9B and 9C ) ., Further , we used previously described statistical analysis 10 to demonstrate that the LTR-gag amplicons from B and NK cells were unlikely to have come from contaminating CD8+ T cells p<0 . 05 ( 1 ) ., The identification and characterization of cell types harboring HIV genomes is crucial for the development of strategies to promote clearance ., HSPCs support both active and latent infection by HIV in vitro and in vivo 13 , 18 ., However , prior studies suggested a model in which only CXCR4-tropic viruses , which infect long-lived HSCs would be capable of persisting in vivo 3 ., Here , we provide evidence that non-stem cell CD34+ progenitors infected by CCR5-tropic viruses are also long-lived ., Indeed , HIV provirus isolated from HIV-infected people treated with cART for years was often CCR5-tropic and recoverable from HSPC populations that were depleted for stem cells ., These unexpected results support recent studies showing that non-stem cell progenitors can persist in vivo for years and provide evidence that they may form a significant reservoir in HIV infected people ., We also provide strong evidence that progenitor cells , including multipotent progenitors , harbor HIV receptors ., These results are consistent with other studies investigating the lineage potential of CD4 subsets using functional assays 16 , 17 , 19 ., Two studies showed that CD34+ CD4high and CD4low/- populations include clonogenic progenitors and Louache et al furthermore demonstrated that CD34+ CD4+ HSPCs are enriched for long-term culture-initiating cells 16 , 17 ., Another study extended these results using human fetal liver to show that CD34+CD4+ cells are able to engraft in an immunodeficient mouse , unlike CD34+CD4- cells 19 ., In addition , HIVs that require CD4 for entry are able to infect and express marker genes in HSCs based on a gold standard functional assay ( stable engraftment and generation of all hematopoietic lineages ) 3 ., Thus , CD4 and other HIV receptors are expressed on hematopoietic progenitors ., Preferential infection of the CD4high subset partially explains another study that was unable to detect provirus in HSPCs from infected people 20 ., In this study , flow cytometry was used to isolate Lin−CD34+ CD4- cells , obtaining a mean purity of 76 . 7% that was substantially lower than the samples described here ( mean purity 94 . 1% for Sort 1 and 90 . 3% for Sort 2 ) ., Based on the data presented here , removal of the CD4+ population would have removed the HSPC population most likely to be infected ., In addition , this small study of 8 donors ( 3 initiating therapy during chronic infection and 5 initiating therapy during acute infection ) was underpowered to detect provirus in HSPCs ., The authors estimate that in these 8 patients , if proviral genomes were present , their frequency would be 0 . 0003%–0 . 003% ( upper 95% confidence bounds ) ., Given that 59% of our donors were positive and that the mean frequency of provirus in our cohort was 2 . 4 copies per million cells ( 0 . 0002%; range 18 to < 0 . 8 copies per million cells ) , the small study size and the small number of cells screened provide additional explanations for why this and another similarly powered study 21 were negative ., Importantly , we isolated four near full-length genomes from HSPCs and a detailed analysis of open reading frames and cis-acting elements revealed they are likely to be functional ., However , demonstration of functional virus using viral outgrowth assays will require additional studies using larger cell numbers ., Studies in T cells have shown that only about a tenth of functional virus can be detected in outgrowth assays 22 ., A Poisson analysis using a mean frequency of 3 copies of provirus per million HSPCs with 30% functional based on sequencing suggests 60 million HSPCs will be needed for 95% certainty of detecting one infectious unit ., Given that we obtain about 2 . 5 million HSPCs for each donor from 20 cc of marrow , we would need to dramatically increase our aspiration size to acquire sufficient cells , which would not be easy to accomplish because of patient discomfort ., The low rate of infection in HSPCs likely explains why an earlier study utilizing low numbers of HSPCs ( approximately one million ) yielded negative results in outgrowth assays 21 ., In addition , while we have shown that transcriptionally latent viral genomes in HSPCs can be reactivated by TNFα and histone deacetylase inhibitors in vitro after cell culture 3 , 13 , studies using large cell numbers are needed to determine the optimal strategies to effectively reactivate proviral genomes to promote viral release from fully quiescent HSPCs tested ex vivo ., Nevertheless , the conclusion that HIV indeed infects HSPCs was confirmed by the detection of clonal HIV proviral genomes in differentiated lineages that matched provirus from HSPCs ., Because the differentiated cells were CD4-negative lineages and because the provirus contained signature inactivating deletions , these results can’t be explained by coincident infection ., Moreover , we confirmed the presence of these genomes by isolating cell-associated mRNA containing the same deletion from activated CD4-negative cells ., Further , we showed by phylogenetic analysis that the genomes frequently isolated from CD4-negative lineages formed a unique clonal population within the donor , indicating that contamination from other cell types was an unlikely explanation of our findings ., In sum , the most likely explanation is that these genomes were transmitted to CD4-negative progeny through differentiation of a CD4-positive progenitor ., In addition , we also detected a proviral genome with a unique signature deletion in both HSPC and CD4+ cells indicating that infected HSPCs can also differentiate into CD4+ cells ., In most cases , detection of proviral genomes in CD4-negative lineages was rare with only a small number of proviral genomes detected per million cells screened ., The exception was donor 431000 in which we detected a defective provirus at a higher frequency ( approximately one per 100 , 000 cells screened ) ., Because replication competent virus could disrupt differentiation due to cytopathic effects , it is not surprising that viral spread from differentiating HSPCs would be uncommon with functional virus , occurring at a higher frequency in cells harboring a defective viral genome that might allow normal differentiation to occur ., In addition , we detected proviral genomes more often in CD4-negative lineages from donors with CXCR4-tropic virus , consistent with its ability to target a wider range of HSPC subtypes , including MPPs and HSCs ., With the exception of one donor ( 431000 ) , we did not find CCR5-tropic provirus in differentiated CD8 , B and NK lineages found in the peripheral blood , which is consistent with observations that CCR5-tropic HIV more commonly infects restricted myeloid progenitor cells 4 ., Although HSCs are the main drivers for reconstitution of all hematopoietic lineages in xenograft models , new insights in animal and human disease models have shown contributions of non-stem cell progenitors to steady state hematopoiesis over long periods of time 6–8 ., Non-stem cell progenitors appear to survive longer than previously thought in the bone marrow without contribution from HSCs , with non-stem cell clones sequentially recruited over time to produce mature blood cells 6–8 , 23 , 24 ., Our data that CCR5-tropic provirus persists for years in non-stem cell progenitors is to our knowledge the first evidence that non-stem cell progenitors persist for years in humans without evidence of bone marrow disease ., Given that non-stem cell progenitors persist , the prevalence of CCR5-tropic HIV in this compartment is not surprising ., During acute infection when circulating virus peaks , the majority of virus is CCR5-tropic 25 ., However , we also detected persistent provirus that encodes Env proteins capable of utilizing CXCR4 to enter cells ., Assuming transmitting virus is nearly uniformly CCR5-tropic , as some studies have indicated , the presence of persistent reservoirs of CXCR4-tropic provirus may indicate that reservoirs continue to form during evolution to CXCR4 tropism in some donors ., Overall , these results support a new model in which non-stem cell progenitors are important long term contributors to normal hematopoiesis and moreover that these cells can serve as a persistent reservoir for HIV provirus ., HIV-infected individuals were recruited through the University of Michigan HIV-AIDS Treatment Program and the Henry Ford Health System ., Written informed consent was obtained according to a protocol approved by the University of Michigan Institutional Review Board and Henry Ford Institutional Review Board ( U-M IRB number HUM00004959 and HFH IRB number 7403 ) ., Donors were >18 years old , with normal white blood cell counts and plasma viral loads were <48 copies/ml for at least 6 months on antiretroviral therapy ., 100 ml of peripheral blood and 20 ml of bone marrow were obtained from each donor ., All collected samples were coded ., Whole umbilical cord blood ( CB ) from uninfected donors was obtained from the New York Blood Center and whole bone marrow was obtained commercially ( AllCells Ltd . ) ., All collected samples were anonymized ., For isolation of HSPCs , mononuclear cells were purified by Ficoll-Hypaque centrifugation and adherence depleted in serum-free StemSpan medium ( StemCell Technologies ) for 1–2 hours at 37°C ., Sort 1 cells were isolated with a CD133 MicroBead Kit ( Miltenyi Biotec ) according to the manufacturer’s protocol , using two sequential sorts for increased purity ., ( For donations 453000 , and 454304 , we used 1 . 5 times the recommended MicroBeads to increase yield . ), Sort 2 cells were isolated from the Sort 1 flow-through using EasySep Human CD34 Positive Selection Kit ( StemCell Technologies ) according to the manufacturer’s protocol , using two sequential sorts ., Where indicated , lineage-positive cells were depleted using the EasySep Lineage Depletion Kit ( StemCell Technologies ) before proceeding to the CD133 magnetic sort ., CD4 negative PBMCs from the human donors described in the ethics statement were purified by depletion of CD4+ cells with MicroBeads ( Miltenyi Biotec ) according to the manufacturer’s protocol modified for a bead:cell ratio of 1 . 5:1 and passage over two sequential LS magnetic columns ., Depleted cells were stained and sorted as indicated in the text to remove residual CD4+ cells on a MoFlo Astrios flow cytometer ., Supernatant and cell associated RNA was extracted using TRIzol LS and TRizol reagents , respectively according to the manufacturer’s protocols ( Invitrogen ) and converted to cDNA using qScript cDNA Supermix or qScript Flex cDNA Kit according to manufacturer’s instructions ( Quanta Biosciences ) ., RNA from viral supernatants was quantified by real time PCR using TaqMan Fast Mastermix ( Applied Biosystems ) on an Applied B

Introduction, Results, Discussion, Materials and methods

Latent HIV infection of long-lived cells is a barrier to viral clearance ., Hematopoietic stem and progenitor cells are a heterogeneous population of cells , some of which are long-lived ., CXCR4-tropic HIVs infect a broad range of HSPC subtypes , including hematopoietic stem cells , which are multi-potent and long-lived ., However , CCR5-tropic HIV infection is limited to more differentiated progenitor cells with life spans that are less well understood ., Consistent with emerging data that restricted progenitor cells can be long-lived , we detected persistent HIV in restricted HSPC populations from optimally treated people ., Further , genotypic and phenotypic analysis of amplified env alleles from donor samples indicated that both CXCR4- and CCR5-tropic viruses persisted in HSPCs ., RNA profiling confirmed expression of HIV receptor RNA in a pattern that was consistent with in vitro and in vivo results ., In addition , we characterized a CD4high HSPC sub-population that was preferentially targeted by a variety of CXCR4- and CCR5-tropic HIVs in vitro ., Finally , we present strong evidence that HIV proviral genomes of both tropisms can be transmitted to CD4-negative daughter cells of multiple lineages in vivo ., In some cases , the transmitted proviral genomes contained signature deletions that inactivated the virus , eliminating the possibility that coincidental infection explains the results ., These data support a model in which both stem and non-stem cell progenitors serve as persistent reservoirs for CXCR4- and CCR5-tropic HIV proviral genomes that can be passed to daughter cells .

People who are effectively treated with antiretroviral medication harbor persistent forms of HIV that are integrated into the cellular genome ., While HIV is cytopathic to most cells , transcriptionally silent , latent forms do not express toxic HIV gene products and can survive in the host for years ., When conditions change , the latent virus can be activated to reinitiate infection ., Because of the capacity for virus to spread , cure of HIV will require that we identify and eradicate all cells harboring functional HIV provirus ., CD4+ T cells are abundant and easily identified as harboring proviral genomes ., However , rare cell types that express HIV receptors , such as bone marrow hematopoietic progenitor and stem cells can also be infected by the virus potentially serving as barriers to cure strategies ., We found that HIV can infect and persist in progenitor sub-types that were previously thought to be short lived , which expands the types of cells that can support reservoir formation ., In addition , we found that HIV can spread by proliferation and cellular differentiation without the need for viral gene expression and virion production that could reveal the infection to the immune system ., A deeper understanding of viral reservoirs is critically important for developing strategies that will succeed in viral eradication .

blood cells, flow cytometry, hiv infections, medicine and health sciences, immune cells, pathology and laboratory medicine, pathogens, immunology, microbiology, cell differentiation, retroviruses, viruses, immunodeficiency viruses, bone marrow cells, developmental biology, rna viruses, stem cells, molecular biology techniques, research and analysis methods, infectious diseases, spectrum analysis techniques, white blood cells, artificial gene amplification and extension, animal cells, medical microbiology, hiv, microbial pathogens, t cells, molecular biology, spectrophotometry, cytophotometry, cell biology, polymerase chain reaction, viral pathogens, biology and life sciences, cellular types, viral diseases, lentivirus, organisms

journal.pgen.1006213

Exploiting Single-Cell Quantitative Data to Map Genetic Variants Having Probabilistic Effects

Modern genetics aims to identify DNA variants contributing to common trait variation between individuals ., A high motivation to map such variants is shared worldwide because many heritable traits relate to social and economical preoccupations , such as human health or agronomical and industrial yields ., In addition to the molecular knowledge they provide , these variants fuel the development of personalized and predictive medicine as well as the improvement of economically-relevant plants , animal breeds or biotechnology materials ., However , this high ambition is accompanied by a major challenge: common traits are under the control of numerous variants that each contribute little to phenotypic variation 1 , and this modest contribution of each variant hampers the statistical power to detect them ., Power is further limited by the multiplicity of linkage tests when scanning whole genomes ., The consequence of this has been debated under the term missing heritability: most of the genetic variants of interest remain to be identified ., Currently , this issue is handled by modelling the effect of known or hidden factors , and by scaling up sample size up to tens of thousands of individuals 2–4 ., Practically , however , cohort size cannot be infinitely increased , and relevant factors are difficult to choose ., Studies would therefore greatly benefit from a better detection of small genetic effects , and from a reduction of the number of genomic loci to test ., Small-effect variants are typically associated with predisposition ( or incomplete penetrance ) : carriers of a mutation display a phenotype at increased frequency , but not all of them do ., In this probabilistic context , the statistical properties of cellular traits may sometimes become informative: a tissue may break because cells have an increased probability to detach , a tumor may emerge because a cell type has an increased probability of somatic mutations , a chemotherapy may fail if cancer cells have an increased probability to be in a persistent state ., In other words , molecular events in one or few cells can have devastating consequences at the multicellular level ., As discussed previously 5 , cellular-scale probabilities are likely related to the genotype and this relation may sometimes underlie genetic predisposition 6 ., Striking examples are genetic factors affecting the mutation rate of somatic divisions and thereby modifying cancer predisposition ., These loci have a probabilistic effect on a cellular trait: the amount of de novo mutations in the cells daughter ., Other loci may modulate the heterogeneity between isogenic cancer cells that underlies tumour progression 7 , 8 and resistance to chemotherapy 9–11 ., They would then change the fraction of problematic cells between individuals and thereby disease progression or treatment outcome ., Fortunately , the experimental throughput of single-cell measurements has recently exploded ., Technological developments in high-throughput flow cytometry 12 , multiplexed mass-cytometry 13 , image content analysis 14–16 and droplet-based single-cell transcriptome profiling 17 , 18 now offer the possibility to estimate empirically the statistical distribution of numerous molecular and cellular single-cell quantitative traits ., We therefore propose to scan genomes for variants that modify single-cell traits in a probabilistic manner , which we call single-cell Probabilistic Trait Loci ( scPTL ) ., This requires to monitor not only the macroscopic trait of many individuals but also a relevant cellular trait in many cells of these individuals ., After scPTL are found , they can constitute a set of candidate loci to be directly tested for a possible small effect on the macroscopic trait of interest ., Methods are needed to detect scPTL ., With its fast generation time , high recombination rate and reduced genome size , the unicellular yeast Saccharomyces cerevisiae offers a powerful experimental framework for developing such methods ., Using this model organism , scPTL were discovered by treating one statistical property of the single-cell trait , such as its variance in the population of cells , as a quantitative trait and by applying Quantitative Trait Locus ( QTL ) mapping to it 19 , 20 ., However , this approach is limited because it is difficult to anticipate a priori which summary statistics must be used ., We present here the development of a genome-scan method that exploits all single-cell values with no prior simplification of the cell population phenotype ., Using simulations and existing single-cell data from yeast , we show that it can detect genetic effects that were missed by conventional linkage analysis ., When applied to a novel experimental dataset , the method detected a locus of the yeast genome where natural polymorphism modifies cell-to-cell variability of the activation of the GAL regulon ., This work shows how single-cell quantitative data can be exploited to detect probabilistic effects of DNA variants ., Our approach is conceptually and methodologically novel in quantitative genetics ., Although we validated it using a unicellular organism , it opens alternative ways to apprehend the genetic predisposition of multicellular organisms to certain complex traits ., We specify here the concepts and definitions that are used in the present study ., Let X be a quantitative trait that can be measured at the level of individual cells ., X is affected by the genotype of the cells and by their environmental context ., However , even for isogenic cells sharing a common , supposedly homogeneous environment , X may differ between the cells ., To describe the values of X among cells sharing a common genotype and environment , we define a single-cell quantitative trait density function f 5 as the function underlying the probability that a cell expresses X at a given level ( Fig 1A ) ., Statistically speaking , f represents the probability density function of the random variable X . In the present study , this function f ( X ) constitutes the phenotype of the individual from whom the cells are studied ., As for any macroscopic phenotype , it can depend on the environmental context of the individual ( diet , age , disease… ) as well as on its genotype ., Single-cell trait density functions also obviously depend on the properties of the cells that are studied , such as their differentiation state or proliferation rate ., We focus here on the effect of the genotype ., Conceptually , cells from one individual may follow a density function of X that is different from the one followed by cells of another individual , because of genotypic differences between the two individuals ( Fig 1B ) ., The important concept is that the genetic difference has probabilistic consequences: it changes the probability that a cell expresses X at a given level , but it does not necessarily change X in most of the cells ., Depending on the nature of trait X and how the two functions differ , such a genetic effect can have implications on macroscopic traits and predisposition to disease 5 ., The term single-cell Probabilistic Trait Locus will refer here to a genetic locus modifying any characteristics of f ( that is , changing allele A in allele B at the locus changes the density function f of X , i . e . fB ≠ fA ) ., A quantitative trait locus ( QTL ) linked to X is a location on a chromosome where a genetic variant changes the mean or the median of X in the cell population ., Similarly , a varQTL is a genetic locus changing the variance of X and a cvQTL is a genetic locus changing the coefficient of variation ( standard deviation divided by the mean , abbreviated CV ) of X in the cell population ., All three types of loci ( QTL , varQTL and cvQTL ) assume a change in f and they are therefore special cases of scPTL ., However , not all scPTL are QTL: many properties of f may change while preserving its mean , median , variance or CV ., The purpose of the present study was to develop an approach that could identify scPTL without knowing a priori how it might change f ., An important question before investing efforts in scPTL mapping is whether genotypes can modify f without affecting its expected value ( the mean of X ) ., If not , then QTL mapping will capture the genetic modifiers of f and searching for more complex scPTL is not justified ., In contrast , if other-than-mean genotypic changes of f are frequent , then scPTL can considerably complement QTL to control single-cell traits ., In this case , scPTL mapping becomes important ., In multicellular organisms , cell types and intermediate differentiation states constitute the predominant source of cellular trait variation ., Studying their single-cell statistical characteristics requires accounting for the developmental status of the cells ., This constitutes a major challenge that can be avoided by studying unicellular organisms ., The yeast S . cerevisiae provides the opportunity to study individual cells that all belong to a single cell type , in the context of a powerful genetic experimental system ., By analysing specific gene expression traits in this organism , we and others identified loci that meet the definition of scPTL but not of QTL 20 , 21 ., This illustrated that , for some traits , scPTL mapping could complement classic quantitative genetics to identify the genetic sources of cellular trait variation ., To estimate if non-QTL scPTL are frequent , we re-analysed an experimental dataset corresponding to the genetic segregation of many single-cell traits in a yeast cross ( Fig 2A ) ., After a round of meiosis involving two unrelated natural backgrounds of S . cerevisiae , individual segregants had been amplified by mitotic ( clonal ) divisions and traits of cellular morphology were acquired by semi-automated fluorescent microscopy and image analysis 22 ., This way , for each of 59 segregants , 220 single-cell traits were measured in about 200 isogenic cells , which enabled QTL mapping of these traits ., We reasoned that if all scPTL of a trait are also QTL , then a high genetic heritability of any property of f should coincide with a high genetic heritability of the expected value of f ., In particular , the coefficient of variation ( CV ) of a single-cell trait should then display high heritability only if the mean value of the trait also does ., To see if this was the case , we computed for each trait the broad-sense genetic heritabilities of both the mean and CV of the trait ., Note that the genetic heritability computed here is not the same as the mitotic heritability of cellular traits transmitted from mother to daughter cells ., Here , a value ( mean or CV ) is computed on a population of cells , and its heritability corresponds to the proportion of its variation that can be attributed to genetic differences between the cell populations ( see methods ) ., Overall , heritability of mean was higher than heritability of CV , and the two types of heritabilities were correlated ( Fig 2B ) ., We also observed that several traits had high heritability of CV and low heritability of their mean value , or vice versa ., This indicates that , for some traits , genetic factors exist that modify the trait CV but not the trait mean ., This observation is in agreement with the complex CV-vs-mean dependency previously reported in this type of data 23 , 24 ., We therefore sought to develop a method that can detect scPTL that do not necessarily correspond to QTL ., One way to identify scPTL from experimental measures is to compute a summary statistic of the trait distribution , such as one of its moments , and then scan for QTL controlling this quantity ., This approach is particularly appropriate when searching for specific genetic effects on f , such as a change in the level of cell-to-cell variability , and a few previous studies successfully used it to map varQTL and cvQTL 19 , 20 , 22 , 25 , 26 ., However , it is less adapted when nothing is known on the way f may depend on genetic factors ., Scanning for scPTL considers the entire distribution of single-cell trait values as the phenotype of interest and searches the genome for a statistical association with any change in the distribution ., We assume that for a set of genotypic categories ( individuals for multicellular organisms , or populations of cells for unicellular ones ) , a cellular trait has been quantified in many individual cells of the same type ., This way , the observed distribution of the trait constitutes the phenotypic measure of individuals ., We also assume that a genetic map is available and the individuals have been genotyped at marker positions on the map ., The method we propose is based on three steps ., First , a distance is computed for all pairs of individuals in order to quantify how much their phenotype differs ., We chose the Kantorovich metric ( also known as the Wasserstein distance or the earth-movers distance ) to measure this distance because , unlike the Kullback-Leibler divergence , it satisfies the conditions of non-negativity , symmetry and triangle inequality and , unlike the Hellinger distance , it does not converge to a finite upper limit when the overlap between distributions diminishes 27 ., The Kantorovich metric can be viewed as the minimum energy required to redistribute one heap of earth ( one f-function ) into another heap ( a second f-function ) ., It has enabled developments in various fields , ranging from mathematics 28 to economy ( the minimal transportation problem ) 29 , 30 to the detection of states from molecular dynamics data 27 ., The next two steps are inspired from methods used in ecology , where spatial distinctions between groups are often searched after determining distances between individuals 31 , 32 ., In step 2 of our method , individuals are placed in a vectorial space while preserving as best as possible the distance between them ( Fig 3A ) ., This is achieved by multi-dimensional scaling , a dimension-reduction algorithm 33 ., The third step is the genetic linkage test itself ., At every genetic marker available , a linear discriminant analysis is performed to interrogate if individuals of different genotypic classes occupy distinct sectors of the phenotypic space ( Fig 3B and 3C ) ., The optimal choice of dimensionality is determined dynamically and a permutation test assesses statistical significance in the context of the corresponding degrees of freedom ., Note that if the dimensions have been reduced to a single one , then canonical analysis is not needed: the phenotypic value of each individual has become a scalar and linkage can be performed by standard QTL mapping ., Finally , scPTL linkage is scored using the Wilks lambda statistics ., Statistical inference is made using empirical p-values produced by permutations where the identities of individuals are re-sampled ., The full procedure is described in details in the methods section ., We first evaluated if our method could detect scPTL from simulated datasets ., To do this , we considered a probabilistic single-cell trait governed by a positive feedback of molecular regulations ., This is representative of the expression level of a gene with positive autoregulation ., As depicted in Fig 4A , the employed model is based on three parameters ., For each individual , a set of parameter values was chosen and single-cell values of expression were generated by stochastic simulations ., We chose to simulate a scPTL that modified the expected values of the parameters so that the skewness of cellular trait distribution is affected ., To do so , we considered a panel of individuals and their genotype at 200 markers evenly spaced every 5cM ., Parameter values of each individual were drawn from Gaussians and the mean of these Gaussians depended on the genotype at the central marker ., This defined two sets of phenotypes that are depicted by blue and red histograms in Fig 4B ., A universal noise term η was added to introduce intra-genotype inter-individual variation which , in real datasets , could originate from limited precision of measurements or from non-genetic biological differences between individuals ., For each of five increasing values of η , about 130 individuals were simulated ., We first scanned the generated dataset by QTL mapping , treating either the mean trait or its variance as the phenotype of interest ., This way , the central scPTL locus was detected only when intra-genotype noise was null or very low ( Fig 4C ) ., This was anticipated because the mean and variance of the simulated trait values slightly differed between the two sets of individuals ., In contrast , our new method allowed to robustly detect the scPTL locus even in the presence of high ( up to 20% ) intra-genotype noise ( Fig 4D and 4E ) ., The results described above using a simulated dataset suggest that the method can complement usual QTL mapping strategies ., To explore if this was also the case when using real experimental data , we applied scPTL scans to the dataset of Nogami et al . 22 mentioned above ( Fig 2A ) where 220 single-cell traits were measured in about 200 cells from segregrants of a yeast cross ., We applied three genome x phenome scans , each one at FDR = 10% ., Two consisted of QTL interval mapping and were done by considering either the mean cellular trait value of the population of cells or the coefficient of variation of the cellular trait as the population-level quantitative trait to be mapped ., The third scan was done using the novel method described here to map scPTL ., Significant linkages obtained from this scan are available in S1 Table ., As shown in Fig 5 , the three methods produced complementary results ., We detected more linkages with the scPTL method than with the 2 QTL scans combined ( 71 vs . 61 traits mapped ) ., This illustrates the efficiency of using the full data ( whole distribution ) of the cell population rather than using a summary statistic ( mean or CV ) ., In addition , we expected that a fraction of scPTL would match QTL , because QTL controlling the mean or CV of cellular traits are specific types of scPTL ., This was indeed the case , with 67% of scPTL corresponding to loci that were detected by at least one of the two QTL scans ., For 11 cellular traits , a locus was found by QTL or cvQTL mapping but it was missed by the scPTL scan ., This illustrates that the methods have different power and sensitivity ., Importantly , 22 cellular traits were associated to scPTL that were not detected by the QTL search , suggesting that some probabilistic effects may affect poorly the traits mean or CV ., Altogether , these observations highlight the complementarity of the different approaches and show that scPTL mapping can improve the detection of genetic variants governing the statistical properties of single-cell quantitative traits ., Examples of scPTL of yeast cellular morphology are shown in Fig 6 ., One of the cellular traits measured was the distance between the center of the mother cell and the brightest point of DNA staining ( Fig 6A ) ., No QTL was found when searching genetic modifiers of the mean or CV of this trait , but a significant scPTL was mapped on chromosome II ., When displaying trait distributions , it was apparent that segregants carrying the BY genotype at the locus had reduced cell-cell variability of the trait as compared to segregants having the RM genotype ( Fig 6A , right panel ) ., Consistently , a small cvQTL peak was seen on chromosome II , although this peak did not reach genome-wide statistical significance ., This trait , which relates to the statistical properties of DNA migration during the early phase of cell division , provided a biological example where scPTL scan identified a genetic modulator of cell-to-cell variability that was missed by the QTL approach ., Three other traits were of particular interest because they mapped to a position on chromosome VIII where a functional SNP was previously characterized in this cross ., This SNP corresponds to a non-synonymous I->S mutation at position 469 in the Gα protein Gpa1p ., It targets a domain that is essential for physical interaction with pheromone receptors Ste2p and Ste3p 34 , 35 ., In the presence of pheromone , Gpa1p is released from the receptor and triggers a signalling cascade of molecular response that causes cell-cycle arrest and cell elongation ( a process called shmooing ) ., In the absence of pheromone , improper binding of Gpa1pI469S to the receptor causes residual activation of the pathway in the BY strain , as seen by transcriptomic profiling 36 , which explains why BY cells are more elongated 24 and proliferate slower 37 than RM cells ., Here we saw that this locus is a scPTL , but not a QTL , of the degree to which cells are elliptical ( Fig 6B ) ., Displaying the distributions of this trait in each segregant revealed a remarkable amount of variability between the segregants , and that the BY allele at the locus corresponded to a modest reduction of the trait value as compared to the RM allele ( sharper mode at slightly lower value ) ., To see if this was due to the GPA1I469S mutation , we examined the data from a BY strain where this mutation was cured 22 ., Remarkably , the single amino-acid substitution caused a mild but statistically significant redistribution of the trait values ( Fig 6B ) ., This change was comparable to the difference seen among the segregants , demonstrating the causality of the GPA1I469S SNP ., Another trait , corresponding to the distance between the bud tip and the short axis of the mother cell , also mapped to this locus , with the RM allele associated to greater cell-cell variability , and data from the GPA1I469S allele-replaced strain validated this SNP as the causal polymorphism ( Fig 6C ) ., These observations suggest that either the residual activation of the pathway in absence of pheromone is not uniform among BY cells , or the proper inactivation of the pathway is not complete in all RM cells ., This , and the fact that the mutation does not prevent BY cells from proliferating ( as compared to pheromone-arrested cells ) , indicate that the detachment of Gpa1pI469S from the receptor is a rare event that has probabilistic effects on the cellular phenotype ., Further investigations based on biochemistry , dynamic recording of individual cells and stochastic modelling are needed to understand how variation in binding affinity accounts for this effect ., The results described here illustrate that scPTL scans can identify individual SNPs that modify single-cell trait distributions without necessarily affecting the trait mean ., Finally , another trait corresponding to the angle of bud site position mapped to two scPTL loci and no QTL ., One of these loci contained the GPA1 gene on chromosome VIII ., Although the phenotype of bud site selection is not related to shmooing , we examined if the GPA1I469S SNP was involved and found that it was not: the allele-replaced strain did not show a different trait distribution than its control ( Fig 6D ) ., Thus , other genetic polymorphisms at the locus should participate to the statistical properties of cellular morphology , by affecting the position of budding sites ., We then explored if scPTL scanning could provide new results when applied to a molecular system that had been extensively characterized by classical genetics ., The system we chose was the yeast GAL regulon which , in addition to be one of the best described regulatory network , presented several advantages ., Natural strains of S . cerevisiae are known to display differences in its regulation 38 , 39 and the transcriptional response of cell populations can be tracked by flow cytometry ., This provides data from large numbers of cells and therefore a good statistical power to compare single-cell trait distributions ., In addition , acquisitions on many genotypes are possible using 96-well plates ., We reasoned that if features of the cell population response segregate in the BY x RM cross ( described above for morphology ) , then scPTL scanning might identify genetic variants having non-deterministic effects on the regulation of GAL genes ., We first compared the dynamics of transcriptional activation of the network in the two strains BY and RM ., This was done by integrating a PGal1-GFP reporter system in the genome of the strains , stimulating them by addition of galactose in the medium , and recording the response by flow cytometry ., As shown in Fig 7 , both strains responded and full activation of the cell population was reached after ~2 hours of induction ., Interestingly , remarkable differences were observed between the two strains regarding the distribution of the cellular response ., The BY strain showed a gradual increase of expression through time that was relatively homogeneous among the cells ( unimodal distribution with relatively low variance ) , whereas the RM strain showed elevated cell-cell heterogeneity at intermediate activation time points ( higher variance , with fraction of non-induced cells ) ., This suggested that genetic polymorphisms between the strains might control the level of heterogeneity of the cellular response at these intermediate time points ., We sought to map one or more of these genetic factors ., To do so , we acquired the response of 60 meiotic segregants of the BY x RM cross ., Using the data collected at each time point , we scanned the genome for scPTL of the reporter gene expression level using the novel genome-scan method described above ., The procedure identified a locus on chromosome V position 350 , 744 that was highly significant ( genome-wide p-value < 0 . 001 ) at 30 minutes post induction , the time at which heterogeneity markedly differed between the BY and RM strains ( Fig 7B and 7C ) ., The locus was also significant at times 20 min ( p < 0 . 005 ) and 40 min ( p < 0 . 005 ) post induction ., Visualizing the distributions of single-cell expression levels at 30 minutes revealed that the RM and BY genotypes at this locus corresponded to high and low cell-cell heterogeneity , respectively ( Fig 7D and 7E ) ., Thus , this locus explains , at least in part , the different levels of heterogeneity observed between the parental strains ., It should therefore also be detected as a varQTL or cvQTL ., This was indeed the case: the LOD score linking the locus to the variance of expression was 4 . 5 and reached statistical significance ( P = 0 . 005 ) ., Importantly , the scPTL was not a QTL: the locus genotype did not correlate with the mean level of expression of the population of cells ( LOD score < 2 . 8 ) ., When surveying the genomic annotations of the locus 40 , we realized that it contained no obvious candidate gene that would explain an effect on the heterogeneity of the response ( such as genes known to participate to the transcriptional response ) ., One potentially causal gene was DOT6 , which encodes a poorly characterized transcription factor that was shown to shuttle periodically between the cytoplasm and nucleus of the cells in standard growth conditions 41 ., Given that, i ) the shuttling frequency of such factors can sometimes drive the response to environmental changes and, ii ) numerous non-synonymous BY/RM genetic polymorphisms were present in the gene , we constructed an allele-replacement strain for DOT6 and tested if the gene was responsible for the scPTL linkage ., This was not the case ., Strains BY and BY-DOT6RM ( isogenic to BY except for the DOT6 gene which was replaced by the RM allele ) displayed very similar transcriptional responses at intermediate times of induction ( S1 Fig ) ., Fine-mapping of the locus and a systematic gene-by-gene analysis are now needed to precisely identify the polymorphisms involved ., By highlighting a novel genetic locus modulating cell-cell variability of the transcriptional response to galactose , our results show that scPTL scanning can provide new knowledge on the fine structure of a well-studied system ., When considering macroscopic phenotypes , it is important to distinguish the situations where scPTL mapping is biologically relevant from those where it is not ., The determinants of human height , for example , act via countless cells , of multiple types , and over a very long period of time ( ~ 16 years ) ., In such cases , the macroscopic trait results from multiple effects that are cumulated and considering the probabilistic individual contribution of specific cells is inappropriate ., Similarly , many tissular traits heavily rely on communications between cells and probabilistic changes in a few may not affect the collective output of the cell population ., In contrast , a number of macroscopic traits can be affected by particular events happening in rare cells or at a very precise time ( see below ) ., In these cases , studying the probabilities of a biological outcome in the relevant cells or of a molecular event within the critical time interval can provide invaluable information on the emergence of the macroscopic phenotype , and scPTL mapping then becomes relevant ., A striking example of such traits is cancer ., Genetic predisposition is conferred by variants affecting somatic mutation rates and these loci are special cases of scPTL: the cellular trait they modify is the amount of de novo mutations in the cells daughter ., These variants have classically been identified by genetic linkage of the macroscopic trait ( disease frequency in families and cohorts ) , and their role on the maintenance of DNA integrity was deduced afterwards by molecular characterizations ., For a review on the genetics of cancer syndrome predisposition , see 42 , 43 ., scPTL mapping is also relevant to the non-genetic heterogeneity of cancer cells which was shown to be associated with tumour progression 7 , 8 and treatment efficiency 9–11 ., Genetic loci changing the fraction of problematic cells are likely modulators of the prognosis ., If the functional properties ( expression level , phosphorylation status , subcellular localization ) of a key molecular player , such as a critical tumor-suppressor gene , can be monitored in numerous individual cells , then scPTL mapping , as presented here , may help identify genetic factors that modulate the activity of this gene in a probabilistic manner ., Once identified , the association of these loci with the macroscopic phenotype can then be tested directly , avoiding at least partly the statistical challenges of whole-genome scans ., To illustrate this , we considered an idealized case where three scales are bridged: at the molecular level , a scPTL affects the expression of a protein X ( same regulation as in Fig 4 ) ; at the cellular level , cells have higher probability to divide if their level of X is low ( Fig 8A ) ; and at the whole-organism level , disease appears if too many cells are present ., Using a stochastic model of this scenario , we simulated a cohort of individuals and recorded the state and number of cells in every individual over time ( Fig 8B , see methods for details ) ., Disease appeared in all individuals , between age 22 and 29 ., Using the data at age 23 , we compared the power of GWAS and scPTL mapping ., For GWAS , the trait of individuals was whether they had declared the disease or not ., For scPTL mapping , the trait was the expression level of X in 10 , 000 of their cells ., As expected by the moderate effect on disease frequency , GWAS failed to detect the locus ( Fig 8C ) ., In contrast , scPTL detection was highly significant from the same cohort of individuals ( Fig 8D ) ., Importantly , although not significant genome-wide , the GWAS score at the locus had a nominal p-value lower than 0 . 01 ( Fig 8C ) ., The locus would therefore be considered significant if it had been the only one tested ., This illustrates the added value of scPTL mapping: while keeping cohort size constant , it can highlight candidate loci of the genome that can then be tested individually for association to the disease ., This power clearly results from, i ) additional traits ( cellular ones ) that are

Introduction, Results, Discussion, Methods

Despite the recent progress in sequencing technologies , genome-wide association studies ( GWAS ) remain limited by a statistical-power issue: many polymorphisms contribute little to common trait variation and therefore escape detection ., The small contribution sometimes corresponds to incomplete penetrance , which may result from probabilistic effects on molecular regulations ., In such cases , genetic mapping may benefit from the wealth of data produced by single-cell technologies ., We present here the development of a novel genetic mapping method that allows to scan genomes for single-cell Probabilistic Trait Loci that modify the statistical properties of cellular-level quantitative traits ., Phenotypic values are acquired on thousands of individual cells , and genetic association is obtained from a multivariate analysis of a matrix of Kantorovich distances ., No prior assumption is required on the mode of action of the genetic loci involved and , by exploiting all single-cell values , the method can reveal non-deterministic effects ., Using both simulations and yeast experimental datasets , we show that it can detect linkages that are missed by classical genetic mapping ., A probabilistic effect of a single SNP on cell shape was detected and validated ., The method also detected a novel locus associated with elevated gene expression noise of the yeast galactose regulon ., Our results illustrate how single-cell technologies can be exploited to improve the genetic dissection of certain common traits ., The method is available as an open source R package called ptlmapper .

Genetic association studies are usually conducted on phenotypes measured at the scale of whole tissues or individuals , and not at the scale of individual cells ., However , some common traits , such as cancer , can result from a minority of cells that adopted a special behavior ., From one individual to another , DNA variants can modify the frequency of such cellular behaviors ., The body of one of the individuals then harbours more misbehaving cells and is therefore predisposed to a macroscopic phenotypic change , such as disease ., Such genetic effects are probabilistic , they contribute little to trait variation at the macroscopic level and therefore largely escape detection in classical studies ., We have developed a novel statistical method that uses single-cell measurements to detect variants of the genome that have non-deterministic effects on cellular traits ., The approach is based on a comparison of distributions of single-cell traits ., We applied it to colonies of yeast cells and showed that it can detect mutations that change cellular morphology or molecular regulations in a probabilistic manner ., This opens the way to study multicellular organisms from a novel angle , by exploiting single-cell technologies to detect genetic variants that predispose to certain diseases or common traits .

chemical compounds, quantitative trait loci, quantitative traits, carbohydrates, galactose, organic compounds, fungi, mathematics, molecular biology techniques, discrete mathematics, combinatorics, research and analysis methods, gene mapping, chemistry, molecular biology, genetic loci, yeast, permutation, organic chemistry, phenotypes, heredity, genetics, monosaccharides, biology and life sciences, physical sciences, organisms

journal.pgen.0030226

Chromosome Structuring Limits Genome Plasticity in Escherichia coli

Genomic analyses have revealed that bacterial genomes are dynamic entities that evolve through various processes , including intrachromosome genetic rearrangements , gene duplication , and gene loss or acquisition by lateral gene transfer 1 ., Nevertheless , comparison of bacterial chromosomes from related genera revealed a conservation of organization 2 ., For example , the genetic maps of E . coli and Salmonella typhimurium that diverged from a common ancestor about 140 million years ago are extensively superimposable 1 ., Multiple forces seem to shape the organization of bacterial chromosomes , and the imprinting of these processes on the chromosome is evident at different levels ., DNA replication initiated at oriC proceeds bidirectionally until the two replication forks meet ., Replication initiation and termination at defined loci result in guanine/cytosine skew between leading and lagging strands due to the mutational differences 3–5 ., In wild-type ( wt ) cells , replication arms coincide with the two compositional skewed halves of the chromosome , hence the name of replichore 6 ., Initiation of replication occurs at oriC , the origin junction of replichores , and in most cases , the two replication forks are predicted to meet at the terminal junction of replichores where skew changes 7 ., Biological processes may exploit these strand-biased sequences defining each replication arm as a target for selection pressure ., Two examples of positive selection at the replichore scale have been well documented in bacteria; first , the octamer χ sequence involved in the RecBCD-mediated recombination process is overrepresented 3 . 5 times in one orientation along each replichore 8 ., Second , FtsK-Orientating-Polar-Sequences ( KOPS ) are overrepresented on one DNA strand ( 9 , see below ) ., Beyond the replichore organization , processes affecting the genome organization at the gene level also shape chromosome structures , and two different parameters might be affected: orientation of gene transcription relative to replication , and location of genes relative to the origin of replication ., Since replication and transcription occur simultaneously on the same DNA molecule , both head-on and co-oriented collisions are thought to occur in replicating bacteria ., It has been originally proposed that highly expressed genes are preferentially positioned on the leading strand to allow faster DNA replication and reduce transcript losses that occur during head-on collisions 10 ., In E . coli , 54% of coding sequences are found on the leading strand , and as for most bacterial species , highly expressed genes such as rRNA operons ( rDNA ) and genes encoding ribosomal proteins are transcribed in the direction of replication ., However , at least in E . coli and Bacillus subtilis , essentiality , not expressiveness , selectively drives the gene-strand bias 11 ., Another parameter thought to shape chromosome structure at the gene level involves the location of genes relative to the replication origin , and gene dosage effect may constrain this positioning ., In fast-growing bacteria , the replication gene dosage effects are mainly associated with the elements of the translation and transcription machinery , i . e . , rDNA , transfer DNA ( tDNA ) , RNA polymerase , and ribosomal protein genes 12 ., In bacteria , selection operates to maintain the two replichores of approximately equal length ., In most cases , the size of the longest replichore corresponds to 50%–60% of the entire chromosome 13 ., In E . coli , the constraint on the size of replication arms is ensured by the presence of ten Ter sites ( TerA–J ) scattered in two oppositely oriented groups in the terminal half of the chromosome ( 14 , Figure 1A ) ., Each of the Ter sites binds Tus , the replication terminator protein , with a specific affinity ., Each replication fork travels across the five Ter sites in the permissive orientation before it encounters a Ter site in the nonpermissive orientation and is blocked ., The forks are thus trapped between oppositely oriented sites , defining a region called the replication fork trap ., In conditions in which Tus blocks replication forks at ectopic Ter sites , creating a region impossible to replicate , the RecBCD pathway of homologous recombination and SOS induction are essential for viability 15–17 ., The need for a high level of homologous recombination protein RecA and helicase UvrD accounts for the requirement of SOS induction for viability 18 ., A detailed study has shown that forks blocked at Ter sites are stable; linear DNA molecules are formed upon arrival of a second round of replication forks and RecBCD-promoted recombination catalyzes the reincorporation of the double-strand DNA ( ds-DNA ) ends made by replication run off 17 ., UvrD was proposed to enable replication forks initiated at recombination intermediates to progress across the Ter–Tus barrier 18 ., Microscopy observations have shown that circular bacterial chromosomes are organized with a particular orientation within growing cells that preserves the linear order of loci on the DNA 19–23 ., The E . coli chromosome consists of four structured macrodomains ( MDs ) and two nonstructured regions 24 , 25 ., The Ori MD containing the origin of replication oriC is centered on migS , a centromere-like structure involved in bipolar positioning of oriC 26 ., The Ori MD is flanked by two nonstructured ( NS ) regions called NSright and NSleft ( Figure 1A ) ., The Ter MD containing the replication fork trap is centered on the terminal replichore junction ., The Ter MD is flanked by two MDs called the Right and Left MDs ., The existence of the four MDs and two NS regions was deduced from genetic data showing that different MDs do not interact during cell growth , but interact with their adjacent NS regions 25 ., Several important processes take place in the Ter MD ., First , replication ends in the Ter MD because of the presence of the replication fork trap ., Second , the replichore junction diametrically opposed to oriC is the region of the change in compositional skew defining the two replichores 7 ., The site-specific recombination site dif is present near the replichore junction and allows the resolution of chromosome dimers into monomers; to be active , dif must be present in a zone of converging KOPS 9 , 27 ., KOPS are recognized by FtsK which translocates the DNA directionally in order to align dif sites at the septum where XerCD can resolve chromosome dimers into monomers ( for review , see 28 , 29 ) ., Third , the Ter MD contains two Non-Divisible Zones ( NDZ ) refractory to inversions ( 30 , see below ) ., Genetic approaches have provided experimental evidence that some chromosome rearrangements are detrimental for growth or , in rare cases , refractory to inversions 30–37 ., Using homologous recombination , intrareplichore inversions ( Intra ) of segments with one endpoint located in the 20%–30% region flanking the terminal replichore junction , i . e . , the periphery of the Ter MD , have been shown to be reproducibly highly problematic or prohibited in E . coli ( for review , see 38 ) ., However , these regions are not refractory to inversions by the site-specific recombination system used here 25 , 37 ., Inversions that split the Ter MD are detrimental for growth and delay cell division 37 ., In a previous study , we have generated strains with chromosomes carrying inverted segments using the λ site-specific recombination system 25 ., Interestingly , we noticed that strains carrying combinations of partner att sites located in the same regions of the chromosome have similar phenotypes upon inversion , and many of the inversions seemed to affect cell physiology ., The results reported here allow us to define extents and limits to plasticity in the E . coli chromosome ., The analysis of detrimental rearrangements allowed the identification of two types of chromosome inversions that , by changing MD organization , severely affect the progression of the cell cycle ., By using the site-specific recombination system of bacteriophage λ , we previously developed a genetic system that allows the construction and detection of genetic inversions in the E . coli chromosome 25 ., We have constructed several series of strains containing one defined att site at a fixed position and its att partner site inserted at random locations; strains carrying combinations of partner att sites that could give rise to viable recombinants have been selected ., Cassettes were designed to detect inversion between att sites: recombination between attL and attR restores lacZ integrity ( Figure 1B ) ., By providing a limiting amount of recombinase , we were able to reveal the existence of MDs that correspond to large regions that are insulated from each other in the cell ( Figure 1A ) ., By providing a high amount of recombinase , recombination between most of the combinations of att sites can be detected , and there is a good correlation between the frequency of collisions and the frequency of recombinants 25 ., There was no correlation between the frequency of inversion and the physiological properties of cells with inverted configuration; inversions occurring at high frequency can be detrimental , whereas those occurring at low frequency can be neutral ( see below ) ., We now analyze in detail the properties of strains carrying chromosomes with the different inverted configurations ., To unravel the consequences of inverting a chromosomal segment on cell physiology , we performed a number of analyses aimed at detecting defects visible at the colony or cell level ., The size of colonies from strains with the inverted chromosome ( lacZ reconstituted , blue colonies ) was compared to that of strains with the wt configuration ( white colonies ) in rich medium ., The effect of these inversions on growth was also measured using a coculture assay in which strains with chromosomes in wt and inverted configurations were compared ( Materials and Methods ) ., To analyze the consequences of the inversion on the nucleoid morphology , cells grown in exponential phase were stained with DAPI , and nucleoids were observed by fluorescence microscopy ( see Materials and Methods ) ., The percentages of cells with different types of nucleoids were numbered according to the cell size ., The number of chromosome origins was estimated by fluorescence-activated cell sorting ( FACS ) analysis ., Viability of strains was tested in different genetic backgrounds affected in pathways related to DNA metabolism ., We used recA mutants and since RecA is required for both homologous recombination and SOS induction , the requirement of SOS induction for viability was tested in lexA ind− ( SOS− Rec+ ) and recA lexAdef ( SOS+ Rec− ) mutants ., When RecA was required , we used mutants affected in the two RecA-dependent recombination pathways , i . e . , RecBC and RecFOR , to identify the pathway involved ., Measurements of SOS induction were performed in a sfiA background to avoid SfiA-dependent filamentation and inhibition of cell division 39 ., SOS induction was quantified in culture by using a plasmid carrying the uidA gene encoding β-glucoronidase under the control of the PsfiA promoter ( see Materials and Methods ) ., In addition , the presence of a plasmid carrying a gfp gene under the control of the PSfiA promoter allowed the direct visualization of the induction of the SOS response at the cellular level ., tus mutants were used to estimate the defects provoked by inverted Ter sites in various configurations ., When recombinant colonies could not be obtained , PCR reactions probing the presence of recombination at the DNA level were used to check for the occurrence of attL–attR inversion and for the presumed lethality conferred by the inversion ., Ten Ter sites are found in the E . coli chromosome , which are bound in vitro by Tus with varying efficiencies 14 ., The Kobs for Tus binding to the very strong TerB site is about 5 × 10−13 M , and the relative arrest activity was estimated to be around 95% ., Although studies were not performed with other Ter sites , the effect of mutations in TerB mimicking the sequence of other Ter sites allows estimation of their respective strength as deduced from Tus–Ter binding affinity measurements and from measure of the replication arrest activity 14 ., TerA–E and TerG are predicted to be very strong sites ( arrest activity greater than 50% ) , TerH moderately strong ( arrest activity around 33% ) , and TerF and TerI–J weak sites ( arrest activity less than 20% ) ., To estimate the respective strength of Ter sites in vivo in a strain producing wt levels of Tus , we have generated strains in which different Ter sites are inverted , and their properties have been analyzed ( Figures S1 and S2 , and Text S1 ) ., Altogether , the results indicate that efficiency of replication arrest at different chromosomal Ter sites correlates with the predictions based on in vitro affinities and on replication arrest activity of TerB mutant sites ., They show that in conditions of wt level of Tus protein , blocking the two forks by the strong TerE and TerA sites renders RecBCD-dependent recombination essential for viability , as previously observed with TerA 15 ., SOS induction is also essential in these conditions ., The effect of the moderate site TerH in the inverted orientation was less severe , but still significant ., Inverted weak TerI and TerJ sites do not appear to affect growth , suggesting they do not significantly impede replication ( Figure S1 and Text S1 ) ., On the E . coli chromosome , the two replichores are of similar size , suggesting that most replication forks meet within the replication fork trap diametrically opposite to the origin ., To evaluate the requirements for the balance of replication arms , we analyzed strains in which inversion endpoints are in each replication arm , asymmetrically relative to oriC ( interreplichore inversion Inter , Figure 2A , Table 1 ) ., As the inverted region contains oriC , these inversions do not change the orientation of sequences or genes relative to replication ., However , because the two endpoints are at different distances from oriC , the size of the replication arms are modified , one becoming greater and one smaller than 50% of the chromosome ., Imbalance of 5%–10% for replication arms has no effect on colony morphology: the colonies with inverted configurations are similar to those with wt configuration ( Figure 2B , 47% for the short arm and 53% for the long arm ( 47–53 ) in strain Inter R-L3 ( Table 1 ) , and 42–58 in strain Inter R-L5 ( Table 1 ) ) ., The effect of these inversions on growth was also measured using the coculture assay containing strains with either a chromosome in wt or inverted configuration: no defect was associated to this genetic rearrangement as the ratio of inverted to wt cells was close to one after 60 generations ( Figure 2C and unpublished data ) ., The cells and nucleoids of strains with either configuration were not distinguishable ( Figure S3 ) ., When the imbalance reached 15% ( 36–64 in Figure 2B , strain Inter R-NSleft1 in Table 1 ) , some defects became apparent ., The recombinant colonies were smaller than noninverted colonies , and the ratio of inverted to wt cells after 60 generations was affected ( 0 . 13 ± 0 . 02 ) , but the cells and nucleoids of the inverted configuration were similar to those of the wt configuration: only 2% of the cells appeared abnormal ( Figure 2D ) ., Around 20% of imbalance ( 30–70 in Figure 2B , strain Inter R-NSleft2 ) , the size of colonies carrying the inversion was affected; in coculture assays , the ratio of cells with inversion to wt configuration was less than 0 . 01 ( Figure 2C ) and microscopic observation showed longer cells with abnormal nucleoids ( 14% of abnormal cells in Inter R-NSleft2 , Figure S3 ) ., Above 20% of imbalance ( 23–77 and 18–82 in Figure 2B , strains Inter R-NSleft4 and Inter R-O1 , respectively , in Table 1 ) , colonies were barely visible , and more than 20% of cells displayed condensed nucleoids , i . e . , a par phenotype , or grew as cells with unsegregated nucleoids ( Figure 2D and Figure S3 , respectively ) ., Interestingly , we noticed that all strains with an imbalance greater than 20% , i . e . , with a replication arm smaller than 30% and the other larger than 70% , were dependent on RecA for viability ( Figure 2B and Table 1 ) ., Recombinant colonies could be obtained in a recFOR background , but not in conditions inhibiting either RecBC DNA recombination or SOS induction , indicating that the RecBC homologous recombination pathway is required for viability in the presence of an imbalance of replication arms greater than 20% ( Table 1 ) ., The dependence on RecA for viability was suppressed by a tus deletion , indicating that the impediment of replication forks by Tus at Ter sites is responsible for lethality in a recA background ( Figure 2B ) ., Finally , a 2- to 4-fold SOS induction was apparent in strains that required recA for viability ( Table 1 ) ., The analysis performed with inverted Ter sites indicated that in cells expressing wt levels of Tus , the replication forks are stopped at the first strong Ter site in the nonpermissive orientation ( 15 and Text S1 ) ., It implies that , when the imbalance is smaller than 20% , the two forks of a same replication round can progress to the replication fork trap ., In contrast , RecBC-dependent recombination is solicited to restart the first fork that reaches a Ter site before the other fork can reach it when the imbalance of replication arms is larger than 20% ., We propose that , in the conditions used , when the shorter replication arm is less than half the longer one , it is fully replicated twice before completion of replication of the longer arm , leading to the formation of DNA double-stranded ends ., These double-stranded ends induce the SOS response and are lethal in the absence of RecARecBC-dependent homologous recombination ., Many natural inversions in bacterial genomes are symmetrical with respect to replication origins and termini ., Scatter plots of the conserved sequences between related species produce an X-shaped pattern , called X-alignment 2 ., These rearrangements reveal that selection operates to maintain replichores of similar lengths; in most genomes , the size of the longest predicted replication arm does not exceed 60% of the chromosome 13 ., By moving the position of the replication fork trap on the genetic map , we have been able to analyze the effect of varying the imbalance of replication arms ., Remarkably , we did not observe negative effects when the imbalance was around 10% , in total agreement with the observed size distribution of replichores in different species ., Some defects appeared when the imbalance reached 15% , and recombinational rescue of replication forks was required above 20% ., The analysis of interreplichore inversions affecting at the same time two MDs revealed that making hybrid MDs while keeping the wt replichore junction unaffected was well tolerated ( Figure S3 ) ., We noticed that for similar levels of imbalance less than 20% , inversions involving endpoints located either in the NS regions or in the Left , Right , and Ter MDs ( Figure S3 and Table, 1 ) behave similarly: the growth of recombinant colonies was slightly affected , and recombinants were viable in a recA background ( Table 1 ) ., It is only when the imbalance exceeded 20% that recombinant colonies were affected and their formation recA-dependent ( Table 1 ) ., Altogether , these results suggest that in the context of interreplichore inversions , the effect of MD disorganization for the Left , Right , and Ter MDs can be well tolerated by the cell ., We noticed that large inversions inside a replichore ( intrareplichore ) with one endpoint in the Ori MD and the other in the NSright region gave rise to recombinants with no strong defects ., Three examples of strains with such rearrangements are shown in Figure 3 ., These inversions encompass 916 , 927 , and 668 kb corresponding to 826 , 828 , and 607 genes , including four , three , and one rDNA operons , respectively ( Figure 3A , strains Intra O-NSright1 to −3 in Table 1 ) ., Similar outcomes were obtained in the left replichore ., For example , the inversion of a 982-kb–long segment that changes the orientation of 942 genes , including two rDNA operons and 34 ribosomal protein genes ( strain Intra L-NSleft1 in Table, 1 ) had no detectable detrimental effects ( Figure 3B and unpublished data ) ., The colonies of strains with the rearranged chromosome had the same size as those with the wt configuration ( Figure 3B ) ., The diagram shown in Figure 3C indicates that even the largest inversion has no detectable effect on nucleoid morphology ., No strong defect was associated with these genetic rearrangements , because the ratio of inverted to wt configuration was above 0 . 75 after 60 generations in coculture assays ( Figure 3D ) ., Finally , colonies carrying these inverted configurations were viable in a recA background , indicating the absence of important DNA damage ( Figure 3B ) ., Altogether , these results indicate that the direction of replication can be inverted through hundred of genes , including rDNA genes , without important consequences for growth ., Furthermore , the results show that inversions between Ori MD and the NS regions are well tolerated ., Similar conclusions were obtained from the analyses of intrareplichore inversions between NS regions and the flanking Right or Left MD in the absence of active Ter sites ( Figure S4 and Table 1 ) ., Therefore , gene orientation , gene dosage , and sequence skews appear to operate only as long-term positive selection determinants ., Our results are in agreement with the evidence 11 , 40 that weakens the proposed influence of replication on gene orientation 41 , 42 ., However , given the large size of bacterial populations , slightly deleterious effects that can be accredited to positioning rDNA and ribosomal protein genes on the lagging strand are most likely sufficient to eliminate such configurations from the population in long-term evolution ., In contrast to well-tolerated inversions described above , two types of intrareplichore inversions were highly detrimental for the cell: the first type involved endpoints located in the Ter and the Right MDs , and provokes the separation of the replication fork trap from the wt replichore junction ., The second type involved inversion between endpoints located in the Ori MD and in the Right MD ., Features of these two detrimental configurations are described in detail below ., Intrareplichore inversions with endpoints in the Right and Ter MDs ( Figure 4A , strains Intra R-T1 to −3 in Table, 2 ) generate a hybrid Right-Ter MD in which the orientation of TerA , TerD , and TerE is modified , creating a replication arms imbalance close to 35%–65% ( see intra R-T1 in Figure 4B ) ., These strains carry two zones of converging KOPS ( Figure 4C ) : the normal one corresponding to the wt replichore junction , and a new one associated with the replication fork trap in the hybrid Right-Ter MD ., Inversion severely affected the growth of colonies ( Figure 4D ) ., The observation of cells with the inverted configuration revealed the occurrence of a high proportion of abnormal cells: 27% of cells showed a par phenotype , 15% formed cells with unsegregated DNA , and 1% of cells were anucleate ( strain Intra R-T1 in Figures 4E , 4F , and S5 ) ., Cells larger than 10 μm with a high amount of nonsegregated nucleoids were observed ., FACS analyses indicated that the number of chromosomes in the large cells ranged from 16 to 32 ( unpublished data ) ., Other strains with intrareplichore inversions between Right and Ter MDs ( Intra R-T2 and −3 in Figure 4 and Table, 2 ) showed the same features ( unpublished data ) ., The origin of the detrimental phenotypes caused by this chromosomal configuration was analyzed by testing different genetic backgrounds ( Figure 4G ) ., It was not possible to obtain viable recombinants in a lexA ind− background , i . e . , in SOS-defective conditions ., SOS induction was directly visualized by the use of a plasmid expressing gfp under the control of PSfiA promoter ( Figures 4F and S5 ) ., Homologous recombination was also required because recombinants with the inverted configuration could not be obtained in a recA , recBC , or in a recA lexAdef background ( i . e . , in conditions of constitutive SOS induction , but in the absence of RecA-dependent recombination ) ., The phenotype and RecA-independence of interreplichore Right-Ter inversions ( Figure S3 , strains Inter R-T1 to −4 in Table, 1 ) suggests that intermingling Right and Ter MDs cannot by itself be responsible for the growth defects of strains Intra R-T1 to −3 in the inverted configuration ., Growth defects and RecA dependence for viability were suppressed in a tus background , indicating that the position of the displaced replication fork trap is responsible for the growth defects ( Figure S5E ) ., The detrimental effects can not originate only from imbalance of replication arms because the imbalance of replication arms is close to 35–65 , a level that does not render RecA essential for viability in interreplichore inversions ( Figure 2 and Table 1 ) ., Three other hypotheses that might account for the growth defects were tested below: the positioning of dif outside of the replication fork trap , the presence of two zones of converging KOPS , and the merging of replication forks far away from the wt replichore junction region ., In these intrareplichore Right-Ter inversions , the replication fork trap is separated from dif ., It was previously reported that the dif site does not need to be present in the replication fork trap to be fully active because the insertion of a ectopic TerA* site near TerA , moving the replication fork trap away from the dif region , did not affect dif activity 43 ., dif is active in any new replichore junction formed after inversion 9 , 27 ., After deletion of dif from its normal position , we reinserted a 28-bp fragment corresponding to dif in the new replication fork trap , in the region where KOPS converge , far away from the wt replichore junction ( Figure 4C , strain Intra R-T2 Δdif difRFT in Table 2 ) ., Strains carrying this inverted configuration still showed strong detrimental defects and were not obtained in a recA background ( Table, 2 ) even though insertion of dif in the new replication fork trap improved nucleoid distribution in a way suggesting dif activity , i . e . , by removing a 12%–15% fraction of filaments ( 47% of abnormal cells instead of 64% in the absence of dif , and 50% when dif is present at its normal location; unpublished data ) ., These results indicate that the absence of dif from the new replication fork trap is not responsible for the observed defects ., These results were corroborated by the viability of Inter Right-Ter inversions in a recA background when dif was deleted ( strain Inter R-T2 Δdif in Table 2 ) , confirming that the RecA dependence for viability of Intra Right-Ter inversions does not result from the lack of dif in the replication fork trap ., To analyze the defects provoked by forming two zones of converging KOPS and by positioning the replication fork trap far from the dif region but without Right and Ter MDs intermingling , we constructed strains with an inversion that positioned the replication fork trap at the limit between the Ter and the Right MDs ( Intra T1 in Figure 4C , and strains Intra T1 and Intra T2 in Table 2 ) ., Colony formation was not affected; cells and nucleoids were similar to those of the wt configuration , and strains carrying inversions were viable in a recA background ( Figure 4C and 4D ) ., These results indicate that as long as sequences belonging to the Ter MD remain together , merging of replication forks far away from the wt replichore junction in the presence of two zones of converging KOPS does not provoke important growth defects ., To determine whether merging of replication forks outside the Ter MD may be responsible for the detrimental effects of intrareplichore Right-Ter inversions , we generated two different genetic inversions in the Right MD ( strain Intra R3 in Figure 4C , and strains Intra R3 and Intra R4 in Table, 2 ) that inverted TerE in a strain in which TerA and TerD are deleted; inversion of the TerE region provoked replication to end in the Right MD , and generated two zones of converging KOPS ( Figure 4C ) and an imbalance of replication arms close to 35–65 ( Table 2 ) ., Recombinant colonies were slightly affected compared to those with a wt configuration ( Figure 4D ) ; cells and nucleoids from both configurations were similar ( unpublished data ) , and recombinants were viable in a recA background ( Figure 4D ) ., These results indicate that replication forks can merge in the Right MD without affecting viability ., Because none of the simple modifications in the chromosome structure can , by itself , account for the growth defect of intrareplichore Right-Ter inversions , we tested whether the defect was dependent on the length of the Right MD that separates the replication fork trap from the wt replichore junction in the Intra R-T1 configuration ., We constructed strain Intra R-T4 ( Figure 4C and Table 2 ) ., In this strain , the chromosome configuration is similar to Intra R-T1 , Intra R-T2 and Intra R-T3 configurations , but the extent of sequences belonging to the Right MD that are embedded in the Ter MD is reduced ( 170 kb compared to 420 kb ) ., Recombinants showed fewer defects; only a fraction of cells ( 13% ) showed a par phenotype , and less than 1% formed cells with unsegregated nucleoids ( Figure S6 ) ., Importantly , strains in the inverted configuration were viable in a recA background ( Figure 4D ) ., These results are in agreement with the hypothesis that the extent of Right MD DNA that separates the Ter region where fork merge from the replichore junction region is responsible for the observed defects ., The combination of the embedding of Ter sequences in the Right MD and finishing replication within these Ter sequences is responsible for the deleterious effect ., The shortening of the region of Right MD that separate the replication fork trap from the wt replichore junction region suppresses the defects ., Together with the observed viability of recA- interreplichore inversions involving the Right and the Ter MDs ( strains Inter R-T1 and R-T2 in Figure S3 and Table 1 ) , these observations support the hypothesis that the requirement of RecA for the viability of intrareplichore Right-Ter inversions results from the separation of the replichore junction region from the region in the Ter MD where replication ends ., It is therefore tempting to speculate that in deleterious configurations resulting from intrareplichore inversion , replication ends in the displaced part of the Ter MD , activities normally associated to the wt replichore junction region cannot be performed , and the cell cycle is affected ., Altogether , these results suggest the existence of a tight temporal and/or spatial coupling between the end of DNA replication in the Ter MD and an unknown activity near the replichore junction region required to progress in the cell cycle ., Further work will be required to determine whether proteins known to function near the terminal replichore junction , FtsK 44 and TopoIV 45 , are involved in this coupling ., The second class of detrimental inversions corresponds to intrareplichore inversions that combine Ori MD with the Left or Right MD ., Because most of the inversions between the Left and Ori MDs also induce a high imbalance of replication arms , we focused our study on the inversion between Ori and Right MDs ., The detrimental effects of intermingling Ori and Right MDs were revealed by combining attL and attR sites inserted at various positions in the Right ( 14′ , 17′ , 19′ , and 22′ ) and Ori ( 0 . 7′ , 97′ , 95′ , 94′ , 92′ , 88′ , 87′ , and 86′ ) MDs (

Introduction, Results/Discussion, Materials and Methods

Chromosome organizations of related bacterial genera are well conserved despite a very long divergence period ., We have assessed the forces limiting bacterial genome plasticity in Escherichia coli by measuring the respective effect of altering different parameters , including DNA replication , compositional skew of replichores , coordination of gene expression with DNA replication , replication-associated gene dosage , and chromosome organization into macrodomains ., Chromosomes were rearranged by large inversions ., Changes in the compositional skew of replichores , in the coordination of gene expression with DNA replication or in the replication-associated gene dosage have only a moderate effect on cell physiology because large rearrangements inverting the orientation of several hundred genes inside a replichore are only slightly detrimental ., By contrast , changing the balance between the two replication arms has a more drastic effect , and the recombinational rescue of replication forks is required for cell viability when one of the chromosome arms is less than half than the other one ., Macrodomain organization also appears to be a major factor restricting chromosome plasticity , and two types of inverted configurations severely affect the cell cycle ., First , the disruption of the Ter macrodomain with replication forks merging far from the normal replichore junction provoked chromosome segregation defects ., The second major problematic configurations resulted from inversions between Ori and Right macrodomains , which perturb nucleoid distribution and early steps of cytokinesis ., Consequences for the control of the bacterial cell cycle and for the evolution of bacterial chromosome configuration are discussed .

Genomic analyses have revealed that bacterial genomes are dynamic entities that evolve through various processes including intrachromosome genetic rearrangements , gene duplication , and gene loss or acquisition by gene transfer ., Nevertheless , comparison of bacterial chromosomes from related genera revealed a conservation of genetic organization ., Most bacterial genomes are circular molecules , and DNA replication proceeds bidirectionally from a single origin to an opposite region where replication forks meet ., The replication process imprints the bacterial chromosome because initiation and termination at defined loci result in strand biases due to the mutational differences occurring during leading and lagging strands synthesis ., We analyze the strength of different parameters that may limit genome plasticity ., We show that the preferential positioning of essential genes on the leading strand , the proximity of genes involved in transcription and translation to the origin of replication on the leading strand , and the presence of biased motifs along the replichores operate only as long-term positive selection determinants ., By contrast , selection operates to maintain replication arms of similar lengths ., Finally , we demonstrate that spatial structuring of the chromosome impedes strongly genome plasticity ., Genetic evidence supports the presence of two steps in the cell cycle controlled by the spatial organization of the chromosome .

cell biology, genetics and genomics, microbiology, eubacteria

journal.pgen.1003461

The Histone Demethylase Jarid1b Ensures Faithful Mouse Development by Protecting Developmental Genes from Aberrant H3K4me3

Embryonic development is characterized by a coordinated program of proliferation and differentiation that is tightly regulated by transcription factors and chromatin-associated proteins ., As embryonic cells differentiate , certain genes are activated while others are repressed , resulting in a unique pattern of gene expression in each cell type ., Histone H3 lysine 4 tri-methylation ( H3K4me3 ) localizes to transcription start sites with high levels present at actively transcribed genes 1 , 2 , even though H3K4me3 at promoters is not a definite indication for transcriptional activity 3 ., Methylation of H3K4 is catalyzed by a family of 10 histone methyltransferases in mammals 4 ., Five of these are members of the Trithorax group of proteins that were first described in Drosophila to be required for maintenance of Hox gene expression by counteracting Polycomb-mediated repression ., In Mll1 and Mll2 mutant mice , target genes are properly activated but expression fails to be maintained leading to embryonic lethality 5 , 6 ., In addition , H3K4 histone methyltransferases function in hematopoiesis 7 , 8 and neurogenesis 9 ., H3K4me3 is found in a constant balance with Polycomb-mediated repressive H3K27me3 ., Presence of both H3K4me3 and H3K27me3 at promoters is referred to as bivalency 10 ., The category of bivalent genes is enriched in developmental regulators and is particularly abundant in embryonic stem cells ( ESCs ) that have the potential for several lineage choices 11 ., Moreover , Polycomb proteins repress non-lineage specific gene expression , thereby ensuring developmental potency of embryonic and tissue stem cells during lineage specification , differentiation and development ( reviewed in 12 ) ., Polycomb proteins are classified into two separate complexes referred to as Polycomb repressive complex 2 ( PRC2 ) , which mediates H3K27me3 , and PRC1 , which catalyzes mono-ubiquitylation of H2A ( H2AK119ub1 ) 13 , 14 ., Classical models propose a sequential mechanism in which H3K27me3 creates a binding site for PRC1 leading to further repression 14 , 15 , even though emerging studies suggest that Polycomb function is more complex 16–18 ., While histone methylation was initially viewed as a stable modification , the discovery of histone demethylating enzymes has changed this paradigm 19 ., Demethylation of H3K4me3 is catalyzed by the JARID1 ( KDM5 ) family , which in mammals has four members: JARID1A , JARID1B , JARID1C and JARID1D 20 ., The Drosophila JARID1 homologue LID ( Little imaginal discs ) is required for normal development 21 , and the C . elegans homologue RBR-2 ( retinoblastoma binding protein related 2 ) regulates vulva formation and lifespan 22 , 23 ., Mice mutant for Jarid1a are viable , displaying only mild phenotypes in hematopoiesis and behavior 24 ., A recent report suggests that Jarid1b mutant mice are embryonic lethal between E4 . 5 and E7 . 5 25 ., The molecular mechanisms underlying this phenotype were not addressed ., In contrast , others obtained viable Jarid1b mutant mice 26 ., However , the requirement of Jarid1b for the differentiation of ESCs along the neural lineage 27 , 28 suggests that Jarid1b may function in mouse development ., In humans , JARID1B is highly expressed in several types of cancer , and it was shown to regulate proliferation of breast cancer cells and a slow cycling population of melanoma cells that promotes prolonged tumor growth ( reviewed in 20 ) ., While the role of Jarid1b in mice remains controversial 25 , 26 , an understanding of its in vivo function is essential to direct future studies evaluating JARID1B as a potential drug target in cancer therapy ., Jarid1b expression has been reported in various tissues during mouse embryogenesis whereas its expression becomes restricted in adults 29 ., Here we report the first detailed analysis of the contribution of Jarid1b to mouse development ., We show that Jarid1b is required for the proper development of several neural systems in the mouse and address the mechanisms underlying the observed defects ., To characterize the function of Jarid1b during mouse development , we generated constitutive Jarid1b knockout mice ., Conditionally targeted Jarid1b mice containing a lacZ-Neo-reporter cassette flanked by FRT sites and in which Jarid1b exon 6 is flanked by loxP sites 28 were crossed with mice constitutively expressing Flp and Cre recombinase to obtain Jarid1b+/− mice ., Jarid1b+/− mice were further intercrossed to generate Jarid1b−/− mice ., Instead of the expected 25 percent of knockout mice , we only obtained 9 . 3 percent of adult Jarid1b knockouts ( Figure 1A ) , suggesting that Jarid1b−/− mice are sub-viable ., Analysis of early and late embryos from Jarid1b+/− intercrosses showed expected ratios while an increased number of Jarid1b knockouts was present among pups found dead during the first day after birth ( Figure 1A ) , indicating that this might be the critical time for survival ., We have previously shown that conditional deletion of Jarid1b using this construct in vitro results in complete loss of Jarid1b protein and no generation of truncated or alternatively spliced variants 28 ., Loss of Jarid1b in vivo was confirmed in all Jarid1b−/− embryos tested ( see examples in Figure S1A and S1B ) , indicating that partial survival of Jarid1b knockouts in not due to incomplete deletion ., Moreover , expression of other Jarid1 family members is unchanged both in vitro 28 and in vivo ( Figure S1C ) ., To determine more precisely when Jarid1b−/− pups die , we performed caesarean deliveries and closely monitored the pups ( Figure 1B ) ., While approximately 95 percent of wild-type pups survive , we found that 50 percent of the knockouts die within the first two hours after delivery and another approximately 20 percent die after 14 to 24 hours ( Figure 1C ) ., All pups that survive the first day , develop normally until adulthood ., Interestingly , survival of Jarid1b+/− pups is also slightly , even though not significantly , reduced during the first day ., While most of the Jarid1b knockouts are grossly normal and not generally growth retarded ( Figure 1D ) , we observed an increased incidence of developmental defects like exencephaly and eye defects among Jarid1b knockouts ( Figure 1B , 1E and 1F ) ., Taken together , loss of Jarid1b leads to major neonatal lethality of which only a small fraction can be explained by severe morphological abnormalities ., There is a large spectrum of physiological systems whose defects can challenge neonatal survival including those affecting parturition , breathing , suckling and neonatal homeostasis 30 ., The first extrauterine challenge for neonates is breathing and since the majority of Jarid1b−/− pups die immediately after birth , we studied the respiratory system in more detail ., Analysis of lungs from E18 . 5 fetuses revealed a normal size and weight ( 3 . 34±0 . 26 versus 3 . 45±0 . 44 percent body weight in heterozygotes versus knockouts , respectively ) as well as a normal lobulation pattern ( data not shown ) ., Next , we isolated lungs from Jarid1b−/− newborns that had died within 2 hours after delivery and had either not shown any sign of breathing or exhibited gasping respiration ( Figure 2A ) ., While the wild-type lung showed saccular inflation , knockout lungs were compact and poorly inflated visible both from gross appearance and histology ( Figure 2B and 2C ) , suggesting that Jarid1b−/− neonates die due to an inability to establish normal breathing ., Moreover , preterm ( E18 . 5 ) Jarid1b−/− lungs were abnormally compact compared to controls ( Figure 2D ) , which might indicate a failure of prenatal breathing activity 31 ., Respiratory failure might be caused by delayed lung maturation characterized by reduced surfactant expression 32 ., Therefore , we analyzed expression of surfactant proteins ( Sftpa1 , Sftpb , Sftpc and Sftpd ) in Jarid1b knockout mice ( Figure S2A ) ., None of the four surfactants was reduced in Jarid1b knockout lungs at E18 . 5 , suggesting that respiratory failure is not due to pulmonary immaturity ., In agreement with this , intrauterine administration of dexamethasone , a glucocorticoid that induces fetal lung maturation 33 , did not improve survival of Jarid1b knockout pups ( Figure S2B ) ., We also examined other physiological systems that are required for neonatal survival including the rib cage , diaphragm , craniofacial appearance and the palate as well as the cardiovascular system 30 , but did not detect any abnormalities in the Jarid1b knockouts ( Figure S3A–S3E ) ., We conclude that while the lungs , skeletal and cardiovascular systems are properly developed , Jarid1b−/− neonates are unable to reliably establish respiratory function ., Immediate breathing after birth is also dependent on brainstem rhythmogenic and pattern forming neural circuits that develop before birth 34 ., We therefore isolated brains from neonates after caesarean delivery , but found no gross abnormalities or differences in size of Jarid1b−/− brains compared to controls ( Figure S3F and S3G ) ., Essential rhythmogenic networks regulating breathing are located in the brainstem ., Therefore , we recorded spontaneous C3–C5 nerve activity in an in vitro brainstem-spinal cord preparation from E18 . 5 embryos ., Surprisingly , given the respiratory defects in newborn Jarid1b knockouts , central respiratory rhythmogenesis was unperturbed in Jarid1b−/− embryos ( Figure S3H ) ., To monitor neurological reflexes of newborn Jarid1b−/− pups , we tested their response to pinching stimuli 35 ., As opposed to control neonates , Jarid1b mutants only weakly reacted to a tail pinch ( Figure S3I ) , suggesting that Jarid1b newborns show motosensory deficits characterized by hyporesponsiveness ., These results together with our previous in vitro data showing that Jarid1b is required for the differentiation of ESCs along the neural lineage 28 prompted us to analyze the development of neural systems in more detail in Jarid1b−/− embryos ., As a first step we analyzed cranial nerves , a pair of 12 nerves that are essential for sensory and motor functions and reside in the mid- and hindbrain 36 ., Defects in cranial nerve development may compromise neonatal survival ., Cranial and spinal nerves can be visualized by whole-mount immunostaining at E10 . 5 using an anti-neurofilament antibody ., Comparison of Jarid1b−/− embryos with controls revealed that while all nerve pairs are present , several cranial and spinal nerves are dysmorphic in the Jarid1b knockouts ( Figure 3A ) ., We used an arbitrary scoring system to quantify the differences between genotypes and found that Jarid1b knockouts are significantly affected while slight defects are already detectable in heterozygotes compared to wild-type ( Figure 3B ) ., Cranial nerves are involved in a diverse range of functions including movement of the eye , innervation of muscles of mastication , facial expression and tongue , and in transmitting information from chemoreceptors to the respiratory center 36 , 37 , and thus , defects in cranial nerve development may be relevant to reduced survival of Jarid1b knockouts ., For example , the hypoglossal nerve ( XII ) , which is dysmorphic in Jarid1b knockouts , innervates the muscles of the tongue , crucial for upper airway aperture during breathing ., Next , we analyzed Jarid1b expression during the time of mouse development when cranial nerves are specified ., From embryonic day 8 , the hindbrain becomes transiently partitioned along the anterior-posterior ( AP ) axis in a series of 8 rhombomeres that influence the spatial distribution of neuronal types 34 ., Using the lacZ-Neo-reporter cassette present in the targeting construct 28 , we observed high ubiquitous expression of Jarid1b in embryonic but not extraembryonic tissues at E8 . 5 ( Figure S4 ) ., Moreover , in agreement with previous reports 38 , at E12 . 5 and E14 . 5 , Jarid1b expression was observed in several neural tissues including the fore- and hindbrain , neural retina , spinal cord and dorsal root ganglia as well as other tissues ( Figures S5 and S6 ) , indicating that Jarid1b could be involved in the development of several organs ., Cranial nerve development is imparted by genes involved in AP patterning and rhombomere specification , neuronal determination or survival and axonal migration 37 ., Compartmentalization of the hindbrain , and in particular rhombomeres 3 and 4 , have emerged as territories for the maintenance of breathing frequency after birth 34 ., Rhombomeres are characterized by specific patterns of Hox gene , Krox20 ( Egr2 ) and Kreisler ( Mafb ) expression , leading us to analyze expression of these genes by RNA in situ hybridization in Jarid1b−/− embryos ., However , we did not observe any defects in the hindbrain patterning of E8 . 75 embryos ( Figure 3C ) , suggesting that other mechanisms are responsible for spinal nerve abnormalities in Jarid1b−/− embryos ., In addition to sporadic cases of exencephaly , we frequently observed defects in eye development in Jarid1b−/− embryos and pups ( Figure 4A–4D ) ., In the most severe cases , eyes were completely absent ( anophthalmia; Figure 4C ) ., Other embryos exhibited microphthalmia ( Figure 4B and 4D ) or an incomplete closure of the optic fissure ( Figure 4A and 4D ) ., Moreover , after birth , the eyelid was often found open in Jarid1b−/− pups while it was closed in control mice at this time ( Figure 4C ) ., Altogether , externally visible eye defects were observed in approximately 22 percent of Jarid1b−/− embryos and pups ( Figure 4E ) , but never in the Jarid1b knockouts that survive to adulthood ., Histological analysis of two microphthalmic Jarid1b−/− eyes at E18 . 5 revealed a misfolding of the neural retina and a much smaller lense ( Figure 4F and 4G ) ., To test whether Jarid1b is expressed in the developing eye , we performed β-galactosidase stainings on sections of E12 . 5 and E14 . 5 eyes from targeted Jarid1b embryos ( Figure 4H ) ., At both stages , Jarid1b is specifically expressed in the inner layer of the neural retina , which contains retinal ganglion cells ., Thus , Jarid1b seems required for the proper development of a mouse neurosensory organ , the eye ., We have previously shown that Jarid1b binds to the transcription start sites of many developmental regulators in mouse ESCs , many of which are also bound by Polycomb group proteins 28 ., Therefore , we speculated that Jarid1b might also regulate Polycomb target genes in vivo ., Hox genes represent classical Polycomb targets and their misexpression in Polycomb mouse mutants results in transformations of the axial skeleton 39 , 40 ., To investigate whether such transformations are also present in Jarid1b mutants , we stained skeletal preparations of E17 . 5 embryos to visualize cartilage and bone ., While we did not observe any defects in the anterior region of the vertebral column ( occipito-cervico-thoracic region ) , we found a transformation of the 26th vertebra , which is supposed to be the last lumbar vertebra ( L6 ) into the first sacral vertebrae ( S1 ) ( Figure 5A and 5B ) ., Moreover , we also observed a transformation of the 34th vertebra ( Figure 5B and Figure S7 ) ., Thus , Jarid1b−/− embryos display posterior transformations of the skeleton , which similar to Polycomb mutants are not completely penetrant 39 , 40 ., To identify genes in addition to the Hox genes that might be misregulated in Jarid1b−/− embryos , we focused on an early embryonic stage ( E8 . 5 ) where morphological defects were not yet observed ., We expected that several of the phenotypes observed in the Jarid1b mutants arise from misspecification events early in development , as genes involved in eye specification , neural tube closure and hindbrain patterning start to be expressed from E8 . 0 41 , 42 ., First , we performed chromatin immunoprecipitation ( ChIP ) followed by sequencing ( seq ) of head regions of E8 . 5 embryos ( Figure S8A ) for H3K4me3 and H3K27me3 to identify genes that change their chromatin state and thus might become misregulated in Jarid1b−/− embryos ., By this analysis , we identified 492 peaks with increased H3K4me3 levels in Jarid1b knockouts versus heterozygotes , whereas only 27 peaks were detected in the reverse comparison ( Figure S8B ) ., Representative examples of loci with increased H3K4me3 in the knockouts as well as loci with unchanged chromatin states are shown in Figure 6A and 6B , respectively ., The results were validated in an independent experiment by ChIP-qPCR showing that the differences in H3K4me3 are reproducible ( Figure 6C ) ., Comparison of genes with increased H3K4me3 in knockout embryos with all genes revealed an enrichment of repressed ( H3K27me3 positive ) and bivalent ( H3K4me3/H3K27me3 positive ) genes among genes with increased H3K4me3 ( Figure 6D and Figure S8C ) , suggesting that aberrant active histone marks accumulate mainly at genes that are usually not actively transcribed ., Gene ontology analysis of genes with increased H3K4me3 in the Jarid1b−/− embryos identified regulators of transcription and development including genes involved in ectoderm , nervous system and skeletal development as significantly overrepresented ( Figure 6E and Figure S8D ) ., To identify genes that are directly bound and regulated by Jarid1b , we also attempted ChIP experiments for Jarid1b in E8 . 5 embryos but unfortunately the results were of low quality due to very limited amounts of starting material ., Instead , we compared genes with elevated H3K4me3 in Jarid1b−/− embryos with genes bound by Jarid1b in ESCs 28 and found that approximately one quarter was bound by Jarid1b in ESCs ( Figure S8E ) ., Thus , it is likely that some of the genes with increased H3K4me3 are also Jarid1b targets during early mouse development ., Next , we performed gene expression analysis of mRNA isolated from E8 . 5 Jarid1b heterozygotes and knockouts ., Except for Jarid1b , we did not identify any genes that were more than 2-fold changed in the knockouts ( Figure S8F ) ., We validated a number of genes by RT-qPCR and confirmed that Jarid1b was not expressed in the knockouts , whereas Jarid1a and Jarid1c as well as L1cam and Pax2 remained unchanged ( Figure S8G ) ., Taken together , while we detected increased levels of H3K4me3 at a number of developmental regulators early in embryogenesis , these chromatin changes do not translate into detectable global transcriptional changes at this stage of development ., Deletion of Jarid1b in ESCs leads to a global increase in H3K4me3 , while global H3K4me3 levels remain unchanged in Jarid1b depleted neural stem cells isolated from E12 . 5 embryos 28 ., Likewise , depletion of JARID1B in MCF7 cells 43 or depletion of Jarid1a in mouse embryonic fibroblasts 24 did not result in a global elevation of H3K4me3 ., To analyze the effect of Jarid1b depletion in vivo , we prepared protein extracts from different stages of embryos ., We confirmed lack of Jarid1b protein in all knockout embryos analyzed ( Figure 7A ) ., While we detected little change in H3K4me3 by immunoblotting in heads of E12 . 5 ( data not shown ) and E14 . 5 Jarid1b−/− embryos , global H3K4me3 levels were strongly increased in heads of late ( E17 . 5 ) Jarid1b−/− embryos and in forebrains of Jarid1b−/− newborns ( Figure 7A ) ., These results suggest that H3K4me3 accumulates in Jarid1b knockouts as embryonic development proceeds , while H3K4me3 levels remain fairly constant during normal fetal development ( Figure S9 ) ., Next , we wanted to know at which classes of genes H3K4me3 accumulates in brains of newborn mice ., Since the brain is a complex and heterogeneous organ , we first determined whether Jarid1b expression is limited to specific regions at this stage of brain development ., However , β-galactosidase stainings on sections of brains from newborns revealed high overall expression of Jarid1b ( Figure 7B ) ., In addition , RT-qPCR analysis showed similar expression of Jarid1b in fore- and hindbrain , which is reduced in heterozygotes and lost in knockouts ( Figure 7C ) ., Thus , we divided the brain into forebrain and hindbrain for ChIP experiments and selected a number of genes that represent different chromatin states ( Figure 7D–7G and S10 ) ., We observed increased H3K4me3 at repressed ( H3K27me3-positive ) genes , including Otx2 , Pax9 , HoxB5 and Hesx1 , and at active ( H3K4me3-positive ) genes ( Sema5b ) , but not at unmodified genes in P0 forebrains ., Some bivalent genes , for example Pax6 , showed increased H3K4me3 and slightly reduced H3K27me3 , while others remained unchanged ( e . g . Neurod2 ) ., Similar results were obtained in independent ChIP experiments using forebrain or hindbrain ., These data suggest that H3K4me3 is increased at transcription start sites in late stages of brain development ., To test which genes are directly bound by Jarid1b , we also performed ChIP for Jarid1b ( Figure 7D–7G and Figure S10 ) ., We detected Jarid1b binding at transcription start sites of bivalent ( Pax6 ) and H3K4me3-positive ( Sema5b ) genes , which is in agreement with our previous findings in ESCs 28 ., Moreover , we detected low levels of Jarid1b binding ( 2- to 4-fold above background ) at several of the H3K27me3-positive loci with increased H3K4me3 in the Jarid1b knockouts ( Otx2 , Pax9 , Hoxb5 ) , suggesting that elevated levels of H3K4me3 at many of these loci are due to a loss of direct association of Jarid1b ., To determine whether changes in chromatin modifications are accompanied by differences in expression , we performed RT-qPCR in P0 brains of controls and Jarid1b knockouts ( Figure 7D–7F and Figure S10 ) ., We detected increased levels of the transcription factor Otx2 in forebrains of Jarid1b−/− newborns ., Furthermore , in line with a shifted balance of H3K4me3 versus H3K27me3 , expression of the neural master regulator Pax6 was increased in Jarid1b−/− P0 brains ., In contrast , expression of actively transcribed genes , like Sema5b , was unchanged despite higher levels of H3K4me3 ., We conclude that Jarid1b mutants accumulate higher levels of H3K4me3 and show increased expression of genes important for regulating embryonic development ., Next , we analyzed at which stage between E8 . 5 and P0 changes in gene expression arise in Jarid1b knockouts ., While we did not detect transcriptional changes at E8 . 5 , expression of Otx2 , Pax6 and Sema5b was increased in heads of E12 . 5 Jarid1b knockout embryos compared to controls ( Figure S11A ) ., Since the transcription factor Pax6 controls the balance between neural stem cell ( NSC ) self-renewal and neurogenesis 44 , we tested whether deletion of Jarid1b affected this balance ., Sorting of NSCs and neuronal progenitor cells ( NPs ) from E12 . 5 brains ( Figure S11B ) revealed a slight ( but not significant ) increase in NSCs in Jarid1b knockouts and no change in NPs ., Similarly , global levels of neuron and astrocyte markers remained unchanged in P0 brains ( Figure S9B ) , which is in agreement with normal gross morphology of Jarid1b knockout brains ( Figure S3F ) ., Thus , the detectable changes in gene expression observed in Jarid1b knockout mice does not appear to be a result of abnormal numbers of NSCs or NPs ., Finally , we tested whether increased expression of Otx2 and Pax6 correlated with survival ., However , as shown in Figure S11C , we did not detect a significant difference in expression of Otx2 and Pax6 in brains of newborns that were alive 2 hours after caesarean delivery versus newborns that died immediately ., In contrast , the expression of Otx2 was significantly higher in the adult brain of surviving knockout animals as compared to wild type ( Figure S11D ) , suggesting that transcriptional regulation by Jarid1b is not restricted to embryogenesis only , but affects selected genes rather than global transcription ., Embryonic development is regulated by transcription factors as well as chromatin-mediated processes resulting in tissue-specific gene expression ., Here , we show that the histone demethylase Jarid1b is required for faithful mouse embryonic development ( see model in Figure S12 ) ., Deletion of Jarid1b results in major neonatal lethality caused by an inability of the newborn mice to establish breathing ., Jarid1b mutant embryos display a number of defects related to neural systems , including the misorganization of cranial and spinal nerves as well as increased incidence of exencephaly that might contribute to neonatal lethality ., Respiratory rhythmogenic circuits in the brainstem of Jarid1b mutant embryos appear intact since a spontaneous motor output on cervical nerves was observed under in vitro conditions ., In agreement , Krox20 and Kreisler , essential genes involved in specification of respiratory-related rhombomeres , are also not affected in mutant embryos ., Thus , we speculate that the breathing problems of Jarid1b mutant neonates may stem from either compromised pattern forming circuits controlling airway patency , or an inability of the rhythmic motor output to reach respiratory muscles , caused by defects in cranial and spinal nerve development ., Several other organs important after birth appeared undisturbed ., However , the spectrum of physiological systems required for neonatal survival is large 30 and we cannot exclude that there are other subtle defects that manifest in secondary physiological problems interfering with survival of Jarid1b knockouts ., Previous in vitro studies of ESCs with either reduced 28 or increased 27 levels of Jarid1b have reported a role for Jarid1b during differentiation of ESCs into neurons ., This raises the question of why Jarid1b is specifically required in neural systems ., During embryogenesis , Jarid1b is expressed in several neural organs including the brain , spinal cord and eye , but also in a number of other systems ( this study , 38 ) ., Moreover , in ESCs , Jarid1b is targeted to transcription start sites of genes that regulate development , including genes involved in neurogenesis and ectoderm development 28 ., Thus , a combination of tissue-specific expression and target gene selectivity might explain neural-specific phenotypes ., It should be noted , however , that other systems are also affected by Jarid1b depletion , exemplified by homeotic transformation of the skeleton ( this study ) or slightly reduced expression of meso- and endodermal markers during embryoid body differentiation of ESCs 28 ., Interestingly , knockdown of Jarid1b in the retina of newborn mice leads to abnormal morphology of rod photoreceptor cells and misregulation of rod-expressed genes 45 , supporting a role for Jarid1b in neuronal cells of the eye ., In addition , other Jarid1 family members have reported functions in behavior and/or neurulation , suggesting that these processes are susceptible to changes in H3K4 methylation ., Jarid1a knockout mice display abnormal clasping of the hindlimbs 24 , while mutations of human JARID1C occur in patients with X-linked mental retardation 46 ., Knockout of Jarid1c in the mouse results in embryonic lethality due to defects in neurulation and cardiogenesis 47 ., Taken together , while several Jarid1 family members are involved in the control of neural systems , they may regulate different aspects of development and cannot fully compensate for the absence of other Jarid1 members resulting in gene-specific phenotypes ., To determine how Jarid1b contributes to the regulation of mouse development , we analyzed global as well as gene-specific histone methylation levels in Jarid1b−/− embryos ., Consistent with the previously reported catalytic activity of Jarid1b 23 , 43 , H3K4me3 was increased around transcription start sites of developmental regulators already early during mouse development , particularly at genes that are normally in a repressed or poised state ., At this early stage of development , we could not detect any global changes in gene expression levels , but we cannot exclude that expression of some genes might be affected in a subset of cells as the E8 . 5 mouse embryo is composed of many distinct cell layers ., For example , the development of the eye initiates at E8 . 0 with the evagination of the optic pit from a subset of cells in the diencephalon 41 ., In analogy , increased H3K4me3 at transcription start sites may reflect small increases in many cells of the early embryo or result from large increases in a subset of cells ., Increased H3K4me3 around transcription start sites may render the associated genes more susceptible to later activation , especially during developmental time windows or in specific cell types where additional signaling molecules create a competent transcriptional state ., As embryonic development proceeds , a global increase in H3K4me3 becomes detectable ., Locally , similar to early embryos , increased H3K4me3 is present at the transcription start sites of genes in Jarid1b−/− brains ., Even though altered H3K4me3 per se may not be sufficient to induce transcriptional changes or cell fate conversions 48 , it may have functional consequences when prompt responses to signaling events are required or for the fine control of steady state transcript levels 49 ., Indeed , we detected increased expression of Pax6 and Otx2 , two master regulators of eye and neural development ( reviewed in 50 , 51 ) , in forebrains of Jarid1b−/− pups ., For both Pax6 and Otx2 , it was shown that not only deletion but also overexpression affect eye and neural lineage development 44 , 52 , 53 ., Binding of Jarid1b itself was found at H3K4me3-positive genes ( both active and bivalent ) , which is in agreement with previous ChIP-seq data 28 , 54 ., Interestingly , the overlap between Jarid1b and Polycomb target genes was functionally supported in this study by the observation that Jarid1b knockout embryos show homeotic transformations of the skeleton , which is a hallmark of Polycomb mutant mice ., Moreover , the observation that Jarid1b is bound to several repressed genes that are marked by aberrant H3K4me3 in the knockouts , suggests that Jarid1b is directly required to prevent aberrant accumulation of active chromatin modifications at developmental regulators during embryogenesis ., Most of the phenotypes that we observed in Jarid1b knockouts occurred with incomplete penetrance ., This is not uncommon and has been reported for other histone demethylases 55 but also transcription factors 56 ., The penetrance is often affected by the genetic background of the mice 50 , 55 ., This might also explain the difference in survival of our knockout mice compared to a previous study 25 ., To test this hypothesis , we crossed our Jarid1b mutant mice that were derived on a C57BL/6 background into a mixed C57BL/6/129 genetic background ( Figure S13 ) ., On the mixed background , 40% of knockouts die after birth , compared to 70% on a C57BL/6 background ., Besides , we observed a similar frequency of exencephaly and a reduced response of the newborns to pinching stimuli , while no eye phenotypes were detected on the mixed background ., These results suggest that different genetic background of mice strains used , could partly explain the divergence of obtained results ., Moreover , many disease-causing mutations only have detrimental defects in a subset of individuals , and phenotypic discordance remains even in the absence of genetic and environmental variation ., It was shown that feedback induction of genes with related functions differs across individuals leading to a buffering of stochastic developmental failure through redundancy 57 ., In the Jarid1b mutants , we did not observe upregulation of other Jarid1 members at the transcript level ., However , we cannot exclude that protein levels are increased , or that these proteins are preferentially recruited to Jarid1b target sites to compensate for lack of Jarid1b ., In addition , systematic analysis of transcript levels and their correlation with phenotypes has shown that variability in gene expression underlies incomplete pen

Introduction, Results, Discussion, Materials and Methods

Embryonic development is tightly regulated by transcription factors and chromatin-associated proteins ., H3K4me3 is associated with active transcription and H3K27me3 with gene repression , while the combination of both keeps genes required for development in a plastic state ., Here we show that deletion of the H3K4me2/3 histone demethylase Jarid1b ( Kdm5b/Plu1 ) results in major neonatal lethality due to respiratory failure ., Jarid1b knockout embryos have several neural defects including disorganized cranial nerves , defects in eye development , and increased incidences of exencephaly ., Moreover , in line with an overlap of Jarid1b and Polycomb target genes , Jarid1b knockout embryos display homeotic skeletal transformations typical for Polycomb mutants , supporting a functional interplay between Polycomb proteins and Jarid1b ., To understand how Jarid1b regulates mouse development , we performed a genome-wide analysis of histone modifications , which demonstrated that normally inactive genes encoding developmental regulators acquire aberrant H3K4me3 during early embryogenesis in Jarid1b knockout embryos ., H3K4me3 accumulates as embryonic development proceeds , leading to increased expression of neural master regulators like Pax6 and Otx2 in Jarid1b knockout brains ., Taken together , these results suggest that Jarid1b regulates mouse development by protecting developmental genes from inappropriate acquisition of active histone modifications .

Histone modifications are involved in transcriptional regulation and thus affect cellular identity , differentiation , and development ., We study the histone demethylase Jarid1b ( Kdm5b/Plu1 ) , as it has been reported to be highly expressed in several human cancers and therefore might present a novel target for anti-cancer therapies ., To gain insights into the physiological role of Jarid1b , we have generated a Jarid1b knockout mouse ., We show that loss of Jarid1b affects survival of newborn mice and that Jarid1b is required for the faithful development of several neural organs ., To understand how Jarid1b regulates embryogenesis , we identified genes with increased H3K4me3 at a genome-wide scale as well as Jarid1b target genes during development ., In Jarid1b knockout embryos , master regulators of neural development are expressed at higher levels , underscoring the importance of Jarid1b in transcriptional regulation ., Furthermore , we extend previous reports of overlapping Jarid1b and Polycomb target genes to show the functional relevance of this observation ., Our results provide the first detailed analysis of the role of Jarid1b in normal development and provide a basis for further studies evaluating the contribution of Jarid1b to tumorigenesis .

biology

journal.pcbi.1005525

A systems-level model reveals that 1,2-Propanediol utilization microcompartments enhance pathway flux through intermediate sequestration

Bacterial microcompartments ( MCPs ) are protein-bound intracellular organelles used by Salmonella enterica , Yersinia pestis , Klebsiella spp ., , and other bacteria to spatially organize their metabolism 1–3 ., MCP metabolons allow the growth of these pathogens on carbon and energy sources , such as 1 , 2-propanediol 4 and ethanolamine 5 , that confer a competitive advantage upon invasion of the host gut 6–10 ., MCPs are typically approximately 150 nm in diameter , with multiple enzymes localized inside a porous , monolayer shell composed of several distinct proteins 4 , 11 , 12; a typical bacterial cell contains several MCP structures when in the presence of the appropriate substrate ., Enzymes are localized to the MCP interior through the interactions of N-terminal signal sequences with the inward-facing helices of MCP shell proteins , and potentially through other uncharacterized interactions 13–15 ., Inside the 1 , 2-propanediol utilization ( Pdu ) MCP metabolon , 1 , 2-propanediol metabolism proceeds as follows: the vitamin B12-dependent PduCDE holoenzyme converts 1 , 2-propanediol to propionaldehyde 14 , then propionaldehyde is converted to either 1-propanol by the NADH-dependent PduQ enzyme 16 or to propionyl-coA by the NAD+-dependent PduP enzyme 17 ( Fig 1A ) ., The PduP and PduQ enzymes are thought to cycle a private pool of NAD+/NADH inside the MCP lumen , enforcing a 1-to-1 stoichiometry for the two reactions ., This assumption is reinforced by experimental observations that mutants defective in one of these genes are not rescued by the cytosolic production of homologues 16 , 18 , 19 ., 1-propanol is not used for cell growth in the presence of 1 , 2-PD , but propionyl-CoA can be utilized either as a carbon source or for ATP generation through substrate-level phosphorylation 20 ., For the purposes of this model , we neglect the downstream products and any effects their concentrations might have on the rates of catalysis by PduCDE and PduP/Q , as there is limited experimental evidence in this regard ., Pdu MCPs are elaborate multi-protein structures subject to exquisite regulation , and much investigation has focused on determining the detailed function of the organelles ., Experiments suggest that metabolic pathways are sequestered in the Pdu and ethanolamine utilization ( Eut ) MCPs in order to protect the cell from toxicity associated with aldehyde intermediates 21 , to prevent carbon loss from the metabolic pathway 19 , and to provide a private pool of cofactors for the encapsulated pathways 18 , 19 ., These mechanistic hypotheses are difficult to confirm experimentally , as directly measuring the concentrations of small molecules inside the MCPs in vivo remains a challenge ., Here we build a coupled reaction-diffusion model of the Pdu MCP and use computational and analytic approaches to assess whether the described biological system produces the hypothesized mechanistic behavior ., The use of a mechanistic model of the MCP allows us to examine potential functions and behavior across a wide range of parameters , providing a framework for incorporating experimental observations and guiding experimental design ., The model presented here follows an approach used to investigate the function of a related organelle , carboxysomes , in the CO2 concentrating mechanism of cyanobacteria 22 ., Modeling of the cyanobacterial system indicated that the carboxysome can significantly increase carboxylation efficiency by reducing the rate of unproductive oxygenation by RuBisCO ., The Pdu MCP , however , differs in critical respects ., The first reaction in the model of carboxysome metabolism , catalyzed by carbonic anhydrase , is reversible; the reaction catalyzed by PduCDE , in contrast , is effectively irreversible ., Secondly , the Pdu MCP may play two roles , toxicity mitigation and flux enhancement , while the carboxysome does not function to sequester a toxic intermediate ., The relative importance of these two functions is unclear a priori , and the modeling approach herein allows us to examine the possible trade-offs between the two ., For simplicity , we model the Pdu MCP as a spherical compartment in the center of a radially symmetric spherical cell ., The model includes passive transport of 1 , 2-PD and propionaldehyde across the cell membranes and MCP shell , possible active transport of 1 , 2-PD into the cell , and the action of the PduCDE and PduP/Q enzymes localized within the MCP ( Fig 1B ) ., Parameters were estimated a priori or based on experimental results ., We have developed a numerical simulation for this spherical geometry with localization of enzymes to the MCP ., By making the assumption of constant metabolite concentrations in the MCP lumen , we have also developed a closed-form analytic solution that well approximates the full numerical solution for a broad range of physically relevant parameter values ( S1 Fig; see also Models ) ., The analytic approximation allows for explicit examination of the relationships between different parameters and the mechanisms in the system ., This analytical solution is used throughout the following analysis ., We find that aldehyde sequestration is the key function of the Pdu MCP , and contributes not only to decreasing aldehyde leakage into the cytosol and the growth medium , as is often discussed in the existing literature 21 , but also to greatly increasing flux through the metabolon by increasing the substrate concentration in the vicinity of the relevant enzymes ., Furthermore , we find that active 1 , 2-PD transport across the cell membrane is dispensable at some external 1 , 2-PD concentrations , including the concentrations at which most laboratory experiments are performed , but not at low external 1 , 2-PD concentrations ., This transport activity has been proposed previously , but never experimentally observed 25 ., Finally , while selective MCP membrane permeability is not always required to achieve optimal substrate concentrations , it is often advantageous in this regard ., The qualitative behavior of our model and quantitative fluxes and metabolite concentrations agree well with existing experimental results , without fitting any model parameters to experimental data ( the MCP shell permeability is optimized to maximize flux , but is not fit to data ) ., Additionally , our results suggest several avenues for continuing computational and experimental investigation , including investigations into 1 , 2-PD active transport , direct characterization and detailed simulation of MCP membrane permeability , and analysis of MCP function in chemostatic cultures ., We derive nondimensional equations which are then solved numerically by a finite-difference approach to find the steady-state concentrations in the MCP , and the solutions in the cytosol follow directly ., We solve the spherical finite-difference equations using the ODE15s solver in MATLAB ., Details of the non-dimensionalization can be found in S1 Text , and the corresponding MATLAB code can be found at https://github . com/cjakobson/pduMCPmodel ., If we assume that the concentration gradients in the MCP are small , then the concentrations PMCP and AMCP are approximately constant and the full solution to the reaction-diffusion equations in the MCP and cytosol can be found analytically ., This assumption is tantamount to assuming that the quantity ξ = K M P Q D V C D E R c 2 > > 1 ( see S1 Text ) ; given our assumptions , we estimate that the value of ξ is approximately 104 ., The detailed solution is shown in S1 Text ., In the case when there is no Pdu MCP , we assume that the same number of enzymes are now distributed throughout the cell ., We can thence derive nondimensional equations which can be solved numerically by a finite-difference approach as above ( see S1 Text ) ., In order to assess the function of the Pdu MCP , we compare the performance of the Pdu MCP system to two alternative organizational strategies for the Pdu metabolic enzymes: uniform distribution of the enzymes throughout the cytosol , and co-localization on a scaffold without a diffusion barrier ., We assess the function of each organization strategy by two criteria:, ( i ) maintenance of the cytosolic propionaldehyde concentration below the toxicity limit of 8 mM 21 in Fig 2A , and, ( ii ) saturation of the PduP/Q enzymes with their propionaldehyde substrate in Fig 2B ., Flux through the Pdu metabolon is maximized when the enzymes are saturated ., In each organizational case , we examine the kinetically relevant propionaldehyde concentration ( Fig 2B ) : without MCPs , this is the cytosolic propionaldehyde concentration; with MCPs , this is the propionaldehyde concentration in the MCP ., In all cases , we initially neglect active transport of 1 , 2-PD at the cell membrane ( jc = 0 ) ., The case of enzymes distributed throughout the cytosol provides an assessment of the baseline efficacy of the Pdu pathway without compartmentalization ., As the Michaelis-Menten constant of the PduP/Q enzymes is approximately 15 mM , above the propionaldehyde toxicity limit , it is impossible to both saturate the PduP/Q enzymes and remain below the toxicity limit in the same location ., In fact , if the PduCDE and PduP/Q enzymes are distributed throughout the cytosol , our model suggests that the steady-state propionaldehyde concentration is maintained at 2 . 4 μM ( several orders of magnitude below the 8 mM toxicity limit ) when the external propanediol concentration is 55 mM ( Fig 2A , S4 Fig ) ., In turn , the PduP/Q enzymes , with a KM of 15 mM , are not saturated ( Fig 2B ) ., Another organizational strategy is localization of the relevant enzymes to a scaffold , without a diffusion barrier ., In this case , the propionaldehyde concentration in the vicinity of the PduP/Q enzymes is 2 . 8 μM , higher than if the enzymes are distributed throughout the cytosol ( in which case the kinetically relevant concentration is 2 . 4 μM ) , but still much lower than the saturating concentration of 15 mM ( Fig 2B ) ., When the enzymes are localized in the MCP ( permeability of 10−5 cm/s for 1 , 2-PD and propionaldehyde ) , the PduP/Q enzymes are exposed to a much higher propionaldehyde concentration of 28 mM ( higher than the saturating concentration ) ( Fig 2B ) , while the propionaldehyde concentration in the cytosol is 1 . 2 μM ( Fig 2A ) ., The presence of a diffusion barrier allows the MCP to decouple the cytosolic propionaldehyde concentration ( responsible for toxicity ) from the kinetically relevant propionaldehyde concentration in the vicinity of the PduP/Q enzymes ., Very low nonspecific permeabilities of the diffusion barrier are unfavorable , however: a MCP with very low permeability ( 10−7 cm/s for 1 , 2-PD and propionaldehyde ) maintains a very low cytosolic propionaldehyde concentration of 1 pM , but also a low concentration of propionaldehyde in the MCP of 230 μM ( Fig 2A and 2B ) ., Optimally permeable MCPs are an effective means of decoupling a potentially toxic cytosolic aldehyde concentration from the kinetically relevant aldehyde concentration inside the MCP ., PduP/Q saturation could also be achieved with a very low cell membrane permeability to propionaldehyde , causing an accumulation of propionaldehyde in the cytosol , but at the cost of cytosolic aldehyde concentrations above the toxicity limit ( S5 Fig ) ., In addition , the membrane permeability to propionaldehyde required for this to occur ( 10−7 cm/s ) is dramatically lower than a physiologically reasonable estimate ( 10−3 cm/s ) ., Decoupling PduP/Q saturation from cytosolic propionaldehyde concentration by encapsulation allows significantly greater carbon flux through the MCP metabolon than in the cases of enzyme scaffolding or no organization ( Fig 2C ) ., The flux of propionaldehyde per cell through PduP/Q in the MCP case ( 2 . 97x10−13 μmol/cell-s ) is four orders of magnitude higher than in either the scaffold or no MCP cases ( 8 . 61x10−17 μmol/cell-s and 3 . 58x10−17 μmol/cell-s , respectively ) ., Interestingly , this improvement is due solely to saturation of the PduP/Q enzymes; the flux through the PduCDE enzyme is similar with MCPs ( 6 . 49x10−13 μmol/cell-s ) and without ( 7 . 41x10−13 μmol/cell-s ) ., PduCDE production of aldehyde is sufficient in all four organizational cases , but without a substantial diffusion barrier in the form of the MCP shell , the aldehyde leaks into the cytosol or the extracellular space before it can be utilized by the PduP/Q enzymes ., To quantitatively evaluate our model we estimate the growth rate resulting from the predicted flux through PduP/Q in the MCP case ., Most parameter estimates were made from literature or a priori ( Table 1 ) ; the nonspecific MCP membrane permeability kc was set to the value that resulted in the greatest flux through the PduP/Q enzymes in our model ( Fig 3A ) ., Again , we initially neglect active transport ( jc = 0 ) ., Our model predicts a flux of 2 . 97x10−13 μmol/cell-s for a cell with MCPs , equivalent to 1 . 74x10−5 pg/cell-s , when the external 1 , 2-PD concentration is 55 mM ., Approximately one-half of this flux can be used for cell growth , since 1-propanol is excreted and not used for metabolism in the presence of 1 , 2-PD , so assuming that a bacterial cell has a dry weight of approximately 0 . 3 pg 23 , our model predicts a time of approximately 9 hours for a cell with MCPs to metabolize enough biomass through the Pdu MCP metabolon to accumulate the mass of one daughter cell ( Fig 3B ) ., This value is in good agreement with experimentally measured doubling times for the growth of Salmonella enterica on 55 mM 1 , 2-PD of approximately 5-10 hours 21 ., We believe the model is well suited to address batch-wise experimental results of this kind because a typical experiment measuring the growth of Salmonella on 1 , 2-PD takes place over tens of hours , while our model has characteristic timescales on the order of seconds , at maximum ( Table 2 ) ., The separation of time scales allows us to treat fluxes in the model as at pseudo steady state , representing a single time point in a growth experiment ., This accounts for the good congruence between our model results and the experimental measurements of steady-state Salmonella growth rates on 1 , 2-PD ., Another putative MCP function is the prevention of aldehyde loss into the growth medium ., We quantify this phenomenon in our model as the net flux of propionaldehyde across the cell membrane into the extracellular space ( so-called “propionaldehyde leakage” ) ., This leakage is lower with MCPs than without ( 3 . 52x10−13 μmol/cell-s as compared to 7 . 32x10−13 μmol/cell-s ) , but is the same order of magnitude as the flux through PduP/Q in both cases when the external 1 , 2-PD concentration is 55 mM ( Fig 2D ) ., For the case with MCPs , the flux through PduP/Q is approximately the same as the leakage flux , while leakage is over twice the PduP/Q flux in the no-MCP case ., The flux through PduP/Q , the leakage flux , and the concentrations of propionaldehyde in the cytosol and the MCP , are plotted for a range of external 1 , 2-PD concentrations in Fig 4 ., Fig 4A shows the absolute metabolite concentrations in the MCP ( dashed ) and cytosol ( solid ) in cells with ( green ) and without ( purple ) MCPs and Fig 4B shows the absolute aldehyde leakage ( dashed ) and PduP/Q flux ( solid ) for these two cases ., At high external 1 , 2-PD concentrations , the cytosolic propionaldehyde concentration is comparable with and without MCPs ( Fig 4A ) ., However , the kinetically relevant propionaldehyde concentration in the vicinity of the PduP/Q enzymes is much higher in the MCP when MCPs than in the cytosol without MCPs ., The flux through the PduP/Q enzymes is therefore much higher and the MCP functions primarily for flux enhancement ., At low external 1 , 2-PD concentrations , the cytosolic propionaldehyde concentration is lower with MCPs , resulting in reduced aldehyde leakage into the extracellular space ., Conversely , the kinetically relevant propionaldehyde concentrations in the vicinity of the PduP/Q enzymes are more similar with and without MCPs in the case of low external 1 , 2-PD ( and very low in either case ) , resulting in similar flux through the PduP/Q enzymes ( Fig 4B ) ., In the case of low external 1 , 2-PD concentration , therefore , the relative difference in propionaldehyde leakage into the extracellular space is large , and the MCP functions primarily to reduce aldehyde leakage ., It should be noted that localization to the MCP improves PduP/Q flux by at least 2 orders of magnitude at all external 1 , 2-PD concentrations , even though the flux for both strategies increases with external 1 , 2-PD ., There exists a transition from primarily flux enhancement to primarily aldehyde loss prevention as the external 1 , 2-PD concentration decreases ., At high external 1 , 2-PD concentrations , cells with and without MCPs lose similar fluxes of propionaldehyde to the extracellular space , but cells with MCPs experience much greater flux through PduP/Q; this is due to saturation of the PduP/Q enzymes by the high propionaldehyde concentration inside MCPs ., At low external 1 , 2-PD concentrations , on the other hand , cells with MCPs are more parsimonious with respect to propionaldehyde , but gain a smaller benefit in flux through PduP/Q since the PduP/Q enzymes are not saturated , even with MCPs ( S6 Fig ) ., We next determined the range of external 1 , 2-PD concentrations that saturate the PduP/Q enzymes in the MCP system ., We explored this question using phase space representations of the saturation of the PduCDE and PduP/Q enzymes ( Fig 5 ) ., In each phase space plot , two model parameters are varied and the effect on enzyme saturation is shown ., We illustrate regions of parameter space in which neither enzyme is saturated ( grey ) , only PduCDE is saturated ( orange ) , or both enzymes are saturated ( blue ) ., Saturation of PduP/Q without saturation of PduCDE was not observed ., Also shown in each phase space are isolines illustrating the parameter values for which the cytosolic concentration of propionaldehyde is 10 nM ( 0 . 001% of the toxicity limit ) and 1 uM ( 0 . 1% of the toxicity limit ) , as well as dotted lines indicating the baseline parameter estimates used in the model ( Table 1 ) ., Recall that the toxicity limit for intracellular propionaldehyde is approximately 8 mM ., Phase space representations of this kind are useful because they allow examination of the behavior of the system over a very wide range of parameter space , encompassing the entire range of physically reasonable values for each parameter in question ., In Fig 5A , for instance , the saturation of PduCDE and PduP/Q is examined as a function of the value of the nonspecific MCP membrane permeability kc and the external 1 , 2-PD concentration Pout ., The blue region indicating saturation of both PduCDE and PduP/Q occurs only at high Pout values comparable to or higher than the baseline concentration ., PduCDE alone can be saturated for a wider range of Pout , as indicated by the extent of the orange region ., Interestingly , for a broad range of Pout concentrations , neither enzyme can be saturated no matter the value of the nonspecific MCP permeability kc ., We therefore conclude that PduCDE and PduP/Q can be saturated by adjusting kc for a Pout greater than 50 mM , but for lower Pout concentrations no value of kc achieves enzyme saturation ., We expect that these lower concentrations are relevant for MCP-mediated metabolism in vivo because we observe that a PPdu::gfp reporter is activated for 1 , 2-PD concentrations as low as 55 μM 24 ., Extending this analysis to the other passive diffusion mechanisms considered in the model , we find that PduP/Q can also be saturated for a range of external 1 , 2-PD concentrations by modulating the cell membrane permeability to 1 , 2-PD , k m P , and the cell membrane permeability to propionaldehyde , k m A; but that for each parameter there exists a lower limit of external 1 , 2-PD concentration ( 30 mM and 20 mM , respectively ) below which PduP/Q cannot be saturated by passive mechanisms ( S7A and S7B Fig ) ., The pduF ORF of the S . enterica Pdu operon is a putative membrane protein , and is speculated to encode a 1 , 2-PD transporter 25 ., We therefore explored the possible role of active 1 , 2-PD transport across the cell membrane in the saturation of the PduP/Q enzymes ., Figs 5 and S7 shows phase space representations of PduCDE and PduP/Q saturation with respect to active 1 , 2-PD transport and two passive transport parameters when the external 1 , 2-PD concentration is 55 mM; S8 Fig shows the same analyses when the external 1 , 2-PD concentration is 0 . 5 mM ., We find that active transport of 1 , 2-PD is dispensable at high external 1 , 2-PD concentrations ( i . e . 55 mM ) , but not at lower external 1 , 2-PD concentrations ( i . e . 0 . 5 mM ) ., When the external 1 , 2-PD concentration is 55 mM , the nonspecific MCP membrane permeability , kc , can adopt a value such that both PduCDE and PduP/Q are saturated for any value of the velocity of active 1 , 2-PD transport across the cell membrane , jc ( Fig 5A ) ., The same is true of the cell membrane permeability to 1 , 2-PD , k m P , when the external 1 , 2-PD concentration is 55 mM ( S7C Fig ) ., In contrast , when the external 1 , 2-PD concentration is 0 . 5 mM , there exists a minimum value ( 1 cm/s ) of the velocity of active 1 , 2-PD transport across the cell membrane , jc , below which PduP/Q cannot be saturated solely by modulating the MCP shell passive diffusion parameter kc ( S8A Fig ) ., Similar minima exist at active transport velocities of 2x10−5 cm/s and 0 . 3 cm/s for the cell membrane passive diffusion parameters k m A and k m P ( S8B and S8C Fig ) ., At this lower external concentration of 1 , 2-PD , therefore , active transport of 1 , 2-PD across the cell membrane may play an important role ., Indeed , with high rates of active transport of 6x103 cm/s , PduP/Q can be saturated for an extremely wide range of external 1 , 2-PD concentrations ( S9 Fig ) ., We can further understand these trends by examining the analytical solution to the model ., The relative contribution of active transport to 1 , 2-PD transport across the cell membrane ( as compared to passive diffusion ) is expressed by the quantity λ = 1 + j c k m P . This suggests that active transport can only be significant if j c k m P > 1 ., Furthermore , transport of 1 , 2-PD across the cell membrane only impacts the steady-state 1 , 2-PD concentration in the MCP when λp* ≈ ΓCDE , where λ p * = ( 1 + j c k m P ) P o u t K M C D E represents active and passive transport across the cell membrane and Γ C D E = τ M C P d i f f τ C D E ( τ c e l l t r a n s τ c e l l d i f f + τ M C P t r a n s τ M C P d i f f + 1 3 ρ + 1 3 ) represents the balance between transport and reaction processes ., From the analytical solution in S1 Text , active transport only bears on the solution when j c k m P > 1 and λp* ≈ ΓCDE ., In S7 , S8 and S9 Figs , when the external 1 , 2-PD concentration is high , λp* is large relative to ΓCDE and changes in λ are inconsequential except at large λ values ( when k m P is small ) ., When the external 1 , 2-PD concentration is low , however , as in S4 Fig , λp* ≈ ΓCDEE and active transport is therefore consequential for a large range of k m P . We also observe that active transport only impacts saturation when j c k m P > 1 , as expected from the analytical solution ., Experimental results suggest that the protein membrane surrounding the Pdu MCP might exhibit selective permeability 26 ., We therefore explored under what conditions such selective permeability was advantageous for MCP function ., We first consider the simple case of nonspecific permeability ., We find that an optimal non-selective MCP shell permeability exists with respect to the kinetically relevant propionaldehyde concentration in the vicinity of the PduP/Q enzymes ( Fig 6A ) ., This optimum value ( 10−5 cm/s ) of a single nonspecific permeability kc reflects a tradeoff between 1 , 2-PD entry to the MCP and trapping of propionaldehyde within the MCP ., The non-selective permeability must be high enough for adequate entry of the PduCDE substrate 1 , 2-PD , but low enough to contribute to the accumulation of the PduP/Q substrate propionaldehyde within the organelle ., This can be seen clearly when the permeabilities of the MCP membrane to 1 , 2-PD and propionaldehyde are varied separately ( Fig 6B and 6C ) ., Lower permeability to propionaldehyde is unambiguously advantageous for trapping propionaldehyde in the MCP ., Higher permeability to 1 , 2-PD , k c P , also increases propionaldehyde concentration inside the MCP , until the permeability is sufficient to equalize the cytosolic and MCP 1 , 2-PD concentrations , at which point there is no further improvement ., The presence of an optimal permeability persists when the ratio of k c P to k c A is fixed at 0 . 1 or 10 and the values are varied together , maintaining this ratio ( S10 Fig ) ., Moreover , decreasing kc entails a tradeoff between leakage prevention and flux enhancement , the two aspects of Pdu MCP function ( S11 Fig ) ., At low kc ( less than 10−7 cm/s ) , aldehyde leakage is prevented , but flux is low; near the optimal kc for flux enhancement ( 10−5 cm/s ) , leakage is comparable with and without MCPs ., At high kc ( greater than 1 ) , the system approaches the case of scaffolding , with only a small flux enhancement and no leakage prevention ., We determine the potential benefit of selective permeability by examining the enzyme saturation phase space with respect to the specific permeabilities k c A and k c P ( Fig 6D ) ., The k c A = k c P line , along the diagonal , in this subspace indicates the performance of the system when the MCP permeability is non-selective ., This line passes through the blue region in which PduP/Q is saturated , indicating that selective permeability is not absolutely required for efficient performance ., However , selective permeability permits a broader range of k c A and k c P values to saturate PduP/Q ( Fig 6D ) ., It is interesting to note that MCP permeability to propionaldehyde must be lower than or equal to MCP permeability to 1 , 2-PD ., As observed above , decreasing MCP permeability to propionaldehyde is unambiguously beneficial for flux , while decreasing the MCP permeability to 1 , 2-PD below the MCP permeability to propionaldehyde is detrimental to relative flux ( S11 Fig ) ., It is also important to note , however , that increasing the concentration of propionaldehyde inside the MCP beyond the concentration required to saturate PduP/Q is not beneficial , since there is no increase in flux , but propionaldehyde leakage increases ., Experimental investigations of Pdu MCP function have consistently demonstrated two key phenotypes for strains that express the Pdu enzymes but fail to form MCPs: a slower growth rate and increased propionaldehyde concentration in the media 21 ., Slower growth could be attributable to two phenomena: “passive” growth retardation due to lower carbon flux through the Pdu metabolon , and “active” growth retardation due to the toxic effects of propionaldehyde in the cytosol ., This second form of growth defect is linked to the accompanying observation that strains lacking MCPs have an increased rate of propionaldehyde leakage into the extracellular space ., These two forms of growth retardation cannot be distinguished by an in vivo experiment measuring the growth of cells and the bulk concentrations of the various metabolites ., Our simple model supports the idea that both of these phenotypes contribute to changes in growth rate: we observe that cells with MCPs have both higher flux through PduP/Q and lower cytosolic concentrations of propionaldehyde than cells without MCPs , and that cells with MCPs exhibit lower aldehyde flux into the growth medium than cells without MCPs ., Additionally , experiments indicate similar 1 , 2-PD depletion from the growth medium with and without MCPs , a phenotype that is evident in the comparable PduCDE fluxes predicted by our model with and without MCPs 21 ., Together these observations indicate that our model captures the important principles of Pdu MCP function ., Indeed , we propose that while toxicity reduction likely plays a role in MCP function , flux enhancement is a crucial , and underappreciated , consequence of encapsulation ., This in turn suggests that experiments should explore the relative contributions of flux enhancement and toxicity mitigation in more detail ., For instance , the effects of propionaldehyde toxicity could be characterized by determining the degree to which aldehyde leakage from MCP-defective cells leads to increased formation of covalent adducts to the cellular genome and proteome ., Other enzyme organization strategies , notably enzyme scaffolds based on DNA , RNA , and protein 26–28 , have been shown to increase the flux through enzymatic pathways , and membrane-less organelles and enzyme clusters are also thought to confer kinetic enhancement 29–31 ., While colocalization can generally be expected to enhance flux in many cases , the extent of the enhancement will depend on the kinetics of the enzymes in question and the other cellular processes involved , such as the loss or toxicity of an intermediate ., Indeed , for the Pdu MCP system , we predict only a modest benefit from scaffolding as compared to the benefit derived from encapsulation ., The problem of organization choice awaits a comprehensive analysis of the relative benefits of scaffolding and other strategies for diverse enzyme systems ., Other compartment-based strategies such as the carboxysome also function to increase flux through the enclosed pathway 22 , 32 ., Compartment-based organization differs from the above systems in that it imposes a diffusive barrier that allows the partitioning of intermediates to a subcellular compartment ., The balance between substrate entry and intermediate trapping therefore results in an optimum nonspecific permeability of the diffusive barrier in the carboxysome case ., Notably , we find that such an optimum exists in the Pdu MCP case , as well ., The kinetics of the cascade in the Pdu MCP , however , are fundamentally different from that of the carboxysome , since the first step of the cascade , catalyzed by PduCDE , is irreversible ., This leads to an increasing benefit of selective permeability , accrued due to the accumulation of intermediates , that exceeds what is predicted for the carboxysome 32 ., It is notable that depsite these key differences , we predict a similar overall kinetic enhancement effect for the Pdu MCP system , as compared to these other organization strategies ., This suggests that nature has evolved disparate enzyme organization modes with convergent overall benefits , providing the metabolic engineer with a wide range of potential organizational options for a given heterologous pathway ., Many behaviors in our model depend strongly on external 1 , 2-PD concentration ., Therefore , care must be taken in applying results from experiments conducted at high 1 , 2-PD concentrations ( 55 mM ) in the laboratory to Pdu MCP function inside the host , and in interpreting the results of batch-wise culture experiments in which the external 1 , 2-PD concentration changes during the experiment , decreasing from 55 mM to 1 mM 21 ., We find that MCP function , as quantified by the relative PduP/Q flux and the relative propionaldehyde flux into the extracellular space , changes in response to factors extrinsic to the cell , such as external 1 , 2-PD concentration ., Indeed , these may change during a single experiment: for example , cells without MCPs may be observed to leak slightly more propionaldehyde into the growth medium than cells with MCPs when grown at high external 1 , 2-PD concentrations , but this leakage discrepancy may increase over the course of a batch experiment during which 1 , 2-PD is depleted from the growth medium ., Whether or not PduP/Q can be saturated without active transport also dep

Introduction, Models, Results, Discussion

The spatial organization of metabolism is common to all domains of life ., Enteric and other bacteria use subcellular organelles known as bacterial microcompartments to spatially organize the metabolism of pathogenicity-relevant carbon sources , such as 1 , 2-propanediol ., The organelles are thought to sequester a private cofactor pool , minimize the effects of toxic intermediates , and enhance flux through the encapsulated metabolic pathways ., We develop a mathematical model of the function of the 1 , 2-propanediol utilization microcompartment of Salmonella enterica and use it to analyze the function of the microcompartment organelles in detail ., Our model makes accurate estimates of doubling times based on an optimized compartment shell permeability determined by maximizing metabolic flux in the model ., The compartments function primarily to decouple cytosolic intermediate concentrations from the concentrations in the microcompartment , allowing significant enhancement in pathway flux by the generation of large concentration gradients across the microcompartment shell ., We find that selective permeability of the microcompartment shell is not absolutely necessary , but is often beneficial in establishing this intermediate-trapping function ., Our findings also implicate active transport of the 1 , 2-propanediol substrate under conditions of low external substrate concentration , and we present a mathematical bound , in terms of external 1 , 2-propanediol substrate concentration and diffusive rates , on when active transport of the substrate is advantageous ., By allowing us to predict experimentally inaccessible aspects of microcompartment function , such as intra-microcompartment metabolite concentrations , our model presents avenues for future research and underscores the importance of carefully considering changes in external metabolite concentrations and other conditions during batch cultures ., Our results also suggest that the encapsulation of heterologous pathways in bacterial microcompartments might yield significant benefits for pathway flux , as well as for toxicity mitigation .

Many bacterial species , such as Salmonella enterica ( responsible for over 1 million illnesses per year in the United States ) and Yersinia pestis ( the causative agent of bubonic plague ) , have a suite of unique metabolic capabilities allowing them to proliferate in the hostile environment of the host gut ., Bacterial microcompartments are the subcellular organelles that contain the enzymes responsible for these special metabolic pathways ., In this study , we use a mathematical model to explore the possible reasons why Salmonella enclose the 1 , 2-propanediol utilization metabolic pathway within these sophisticated organelle structures ., Using our model , we can examine experimentally inaccessible aspects of the system and make predictions to be tested in future experiments ., The metabolic benefits that bacteria gain from the microcompartment system may also prove helpful in enhancing bacterial production of fuels , pharmaceuticals , and specialty chemicals .

medicine and health sciences, chemical compounds, pathology and laboratory medicine, enzymes, cell processes, permeability, enzymology, organic compounds, toxicology, toxicity, active transport, materials science, enzyme metabolism, cellular structures and organelles, enzyme chemistry, proteins, chemistry, cell membranes, biochemistry, cell biology, aldehydes, organic chemistry, biology and life sciences, material properties, physical sciences, cytosol

journal.pgen.1006661

Analysis of large versus small dogs reveals three genes on the canine X chromosome associated with body weight, muscling and back fat thickness

Body size variation observed across domestic dog ( Canis lupus familiaris ) breeds provides one of the most visual examples of human selection ., Dogs are thought to have been domesticated about 15 , 000–30 , 000 years ago 1–10 , with the grey wolf being the closest living ancestor 2 , 4 , 11–14 ., The majority of modern dog breeds , however , were developed within the past 300 years with over 340 official breeds noted worldwide 15 , 16 ., The creation of breeds requires codified standards that describe the physical characteristics of the dog ., The breeding strategies used to create dogs with highly specific features have resulted in relatively isolated , pure breeding populations 17 ., The same selective pressures that have reduced phenotypic and genotypic heterogeneity within breeds 6 , 10 , 18–21 result in long stretches of linkage disequilibrium ( LD ) in dogs 20 , 22–24 ., Given these advantageous features , studies of dog breeds have led to the identification of disease genes of interest for human health and biology , including rare human disorders 25–28 , e . g . cancer 29–33 ., The same genomic characteristics have also produced a stellar system for identifying the genes underlying both simple and complex morphologic traits , including coat color and texture variation , tail curl , ear position , skull shape , chondrodysplasia and body size ( reviewed in 34–39 ) ., Body size is the most striking of these traits , as the difference in skeletal size from the smallest to largest dog breeds is about 40-fold 15 , 16 ., Initial studies of dog body size focused on the Portuguese Water Dog ( PWD ) , a breed for which the American Kennel Club ( AKC ) permits about a 50% level of size variation amongst members of the breed 40 ., A genome-wide association study ( GWAS ) of PWD representing a range of body sizes identified the insulin-like growth factor-1 gene ( IGF1 ) 41 , and additional studies of Miniature Poodles and Dachshunds implicated the IGF1 receptor ( IGF1R ) as well 42 , both of which are important regulators of body size ., The IGF1 pathway has also been established as important in normal stature in humans , and mutations in IGF1 have been shown to reduce body size in mice 43–46 ., Four additional positional candidate genes contributing to variation in canine body size have been identified: the Growth Hormone Receptor ( GHR ) on canine chromosome 4 ( CFA4 ) ; High Mobility Group AT-hook 2 ( HMGA2 ) gene on CFA10; Stanniocalcin 2 ( STC2 ) on CFA4; and SMAD family member 2 ( SMAD2 ) on CFA7 22 , 47 , 48 ., The most closely associated variants have been reported for each 47 ., This includes two non-synonymous SNPs in exon 5 of GHR , a SNP in the 5’UTR of HMGA2 , a SNP located 20 kb downstream from STC2 and a 9 . 9 kb deletion 24 kb downstream from SMAD2 , all of which are highly associated with lower Standard Breed Weights ( SBW ) 47 ., Additional studies noted ten additional putative loci 23 , 48 ., These studies did not , however , identify causal variants and did not ascertain the contribution of each gene , or a combination of genes , to overall size variance in dogs ., Variant haplotypes of the six genes described above are strongly associated with large versus small body size , although some exceptions exist ., While we showed previously that IFG1 , IFG1R , SMAD2 , HMGA2 , STC , and GHR variants account for about 60% of body size variance in breeds with a SBW ≤41 kg ( 90 lb ) which are referred hereto as “small/medium breeds , ” the same genes account for <5% of variance in breeds with a SBW >41 kg , hereto referred to as “large breeds” ., We initially identified two loci on the X chromosome spanning several megabases ( Mb ) as contributors to body size in large breeds through a GWAS of 915 dogs representing 80 domestic dog breeds 22 ., The result has since been replicated by several groups 22 , 23 , 47 , 48 ., No study , however has explored the result in detail , in part because the lack of heterozygosity on the canine X chromosome can reflect popular sire effects , which may complicate fine mapping efforts ., In this study we investigate body size loci on the X chromosome using SNP chip data from ≥800 dogs 21 , together with whole genome sequencing ( WGS ) data ., We show that both of the previously identified loci are strongly associated with large breeds , and we perform fine mapping at each locus using WGS data from 163 breeds ., Together , these data reveal associations with three excellent positional candidate genes: Insulin Receptor Substrate 4 ( IRS4 ) which interacts with multiple growth factor receptors such as IGF1R 49 , Immunoglobulin superfamily member 1 ( IGSF1 ) which is involved in the biosynthesis of thyroid hormones 50–52 and Acyl-CoA Synthetase Long-chain family member 4 ( ACSL4 ) which plays a role in lipid biosynthesis and fatty acid degradation 53 ., We initially genotyped a large dataset of 855 dogs representing 88 breeds on the Illumina 170k Canine HD Array 21 ., For purposes of this analysis , large breeds included 165 dogs from the 19 following giant breeds: Akita , Anatolian Shepherd Dog , Bernese Mountain Dog , Black Russian Terrier , Bullmastiff , Dogue de Bordeaux , English Mastiff , Great Dane , Greater Swiss Mountain Dog , Great Pyrenees , Irish Wolfhound , Kuvasz , Leonberger , Neapolitan Mastiff , Newfoundland , Rottweiler , Saint Bernard , Scottish Deerhound , and Tibetan Mastiff ., Using the array data , we compared the genotypes from the above large breeds to 690 dogs from 69 small/medium breeds ( S1 Table ) ., To correct for cryptic relatedness and sex , we used GEMMA 54 , 55 , a linear mixed-model method which accounts for population stratification and relatedness ., A total of 81 SNPs were significant at a genome-wide level for the trait of body mass , and which passed the Bonferroni significance threshold ( -log10 ( P ) >6 . 48 ) ( Fig 1A ) ., Among these , we identified two primary loci on the X chromosome ( Table 1 ) ., The first locus ( locus 1 ) included 23 SNPs , and spanned 82 , 296 , 039 to 84 , 376 , 308 bp ( Fig 1B ) ., A stronger signal ( P = 7 . 74x10-14 ) was identified at a second locus on the X chromosome , which spans 101 , 646 , 292 to 103 , 984 , 352 bp , corresponding to 56 additional SNPs that passed the significance threshold ( Fig 1C ) ., Neither of these loci were within the pseudoautosomal region of the X . Two additional SNPs located on CFA6 passed the significance threshold , chr6: 38 , 284 , 916 ( P = 9 . 36x10-8 ) and chr6: 67 , 350 , 922 ( P = 1 . 80x10-7 ) , but no additional associated SNPs were found in these regions and the result was not explored further at this time ., We examined the loci of interest more closely by calculating pairwise linkage disequilibrium ( LD ) between SNPs within the 4 , 000 kilobase ( kb ) regions surrounding the most strongly associated SNPs at each of the two loci on the X chromosome ., Thirty-five SNPs were highly correlated ( pairwise r2 >0 . 8 ) at locus 1 ( Fig 1B ) while 53 were highly correlated at locus 2 ( Fig 1C ) ., We next investigated each locus by focusing on regions in which SNPs had pairwise r2 values >0 . 5 and extending these regions by +/- 200 kb ., The two refined intervals ranged from 82 , 079 , 576 to 84 , 576 , 308 for locus 1 and from 101 , 378 , 080 to 104 , 418 , 823 for locus 2 ., Locus 1 contains 17 annotated protein-coding genes and 11 annotated RNA genes ( small RNAs and long non-coding RNAs ) or pseudogenes ( S1 Fig ) ., Among these genes , the strongest candidate gene to emerge at locus 1 is Insulin Receptor Substrate 4 ( IRS4 ) , a gene involved in the thyroid hormone pathway , which is associated with IGF1R signaling and body mass index 49 , 56 ., At locus 2 , 20 protein-coding genes are annotated , including a cluster of olfactory receptor genes , and seven noncoding RNAs including microRNA , noncoding RNA and pseudogenes ( S2 Fig ) ., From these 20 genes , we identified one striking candidate , Immunoglobulin Superfamily member 1 ( IGSF1 ) , that encodes an immunoglobulin in the thyroid hormone pathway , and which was previously associated with obesity in IGSF1-deficient humans 50–52 ., To identify functional variants within these two critical intervals , notably in the two strongest candidate genes , IRS4 and IGSF1 , we used WGS data from 163 purebred dogs inclusive of 87 breeds representing the full range of body height and weight specified by the American Kennel Club ( AKC ) ., Each WGS had a mean read depth of at least 10x ( S2 Table ) ., Among these , 21 dogs from 14 breeds were considered large ( SBW >41 kg 90 lb ) , including the English Mastiff , Irish Wolfhound and Saint Bernard ., We first filtered to retain biallelic variants including SNPs and small insertions or deletions <100 bp , with a minor allele frequency ( MAF ) >0 . 05 ., We next screened the remaining biallelic variants , keeping only variants for which the major allele frequency in large breeds was >0 . 5 and <0 . 5 in small/medium breeds ( SBW ≤41 kg 90 lb ) ., A total of 6 , 809 variants remained for locus 1 and 1 , 997 variants for locus 2 ( S3 and S4 Tables ) ., Using these biallelic variant datasets , we performed a new association study for both loci using GEMMA , a linear mixed-model software 54 , 55 , thus defining which alleles were the most strongly associated with large breeds , and in each case that allele was termed the “large allele” ., The 6 , 809 variants identified at locus 1 define a set of genotypes which correspond to a single large haplotype present in more than 90% of large breeds ( S1 Fig ) ., This spans the strong signal originally identified in the GWAS presented here ., Among these variants , we identified one codon deletion ( chrX . g . 82288614-82288616delTCG ) and one insertion ( chrX . g . 82288998-82288999insGCT ) both in the exonic region of the IRS4 gene that were in LD with one another ( S1 Fig ) ., Neither , however , are likely to be significant for this study as neither mutation changes the IRS4 protein size , distinguishes between various size breeds or is in a well-conserved region ( Table 2 ) ., In addition , for each variant the “large alleles” were also identified in more than 20% of small/medium breeds ., While we discarded the above variants in IRS4 from an association with body size , a re-analysis aimed at finding structural variants revealed a large 56 kb deletion ( ChrX:82455513–82511744 ) located 150 kb upstream from the starting codon of IRS4 ( S1 Fig ) ., The variant was only present in the Bernese Mountain Dog , Black Russian Terrier , English Mastiff , Greater Swiss Mountain Dog , Rottweiler , and Saint Bernard ., Visualization of the deletion on an agarose gel indicated that it was also present in multiple other large breeds: the Alaskan Malamute , Bouvier des Flandres , Bullmastiff , Dogo/Presa Canario , Dogue de Bordeaux , Kuvasz and Leonberger ., Among the 6 , 809 biallelic variants identified at locus 1 , we also found three variants , distinct from the above , which were themselves in LD ( Fig 2 ) , and which harbored the highest p-values ( 10−10<P-value<10−15 , P-Wald test ) ( Table 3 ) ., One of the three is a SNP ( chrX . g . 82919525G>A ) in the 5’UTR of Acyl-CoA Synthetase Long-chain family member 4 ( ACSL4 ) , a gene which plays a role in lipid biosynthesis and fatty acid degradation 53 ., This nucleotide is included in a highly conserved region also identified in the human and mouse genomes ( S3 Fig ) ., The other two SNPs were intergenic or intronic ( in AMMECR1 ) ( Table 3 ) ., All three variants , the SNP in ACSL4 , together with the two SNPs in the same LD block , were present in an interesting subset of large dogs ., Specifically , variants were only identified in four of 19 large breeds: Bernese Mountain Dog , Greater Swiss Mountain Dog , Rottweiler , and Saint Bernard ., The other 78 breeds , which included large , medium and small breeds , lacked all three variants ., Of note , all three variants were missing in several large breeds that were skeletally quite large , but comparatively lean , including the Cane Corso , Great Dane , and Irish Wolfhound , among others ( Fig 2 ) ., The breeds in which the variant is found are not simply skeletally large , but also considered “bulky , ” with considerable muscle and fat ., We next checked for the frequency of the 5’UTR ACSL4 variant by testing a larger panel of 959 dogs from 102 breeds , which represented an additional 54 breeds ( S5 Table ) ., The “bulky allele” was present in several dogs with a bulky , heavily muscled body: Bullmastiff , Dogue de Bordeaux , English Mastiff , Greater Swiss Mountain Dog , Newfoundland , Rottweiler and Saint Bernard , where it appears fixed in nearly 100% of dogs from each breed ( Fig 3 ) ., We found that the Alaskan Malamute , Bernese Mountain Dog , Black Russian Terrier , Bouvier des Flandres , Dogo/Presa Canario , Kuvasz and Leonberger breeds could be either heterozygous or homozygous for both alleles ., In total , 48% of the large breeds shared the “bulky allele” ( heterozygous or homozygous ) ( Fig 3 ) ., Sanger sequencing of a larger panel of dogs ( ≥10 dogs per breed ) including the Anatolian Shepherd Dog , Great Dane , Great Pyrenees , Irish Wolfhound , Neapolitan Mastiff , and Scottish Deerhound confirmed the absence of the “bulky allele” in these breeds , many of which are long and lean rather than bulky ., Of note , the ACSL4 variant mutation was never observed in medium or small breeds , even small muscled breeds such as American Staffordshire Terrier , Boston Terrier , or Bulldog ., The results were the same with the two intergenic or intronic variants in LD with the 5’ UTR ACSL4 variant ( S5 Table ) ., Sanger sequencing of the set of wild canids ( 24 grey wolves , two red wolves and two coyotes ) confirmed that the three mutations , including the ACSL4 variant , are absent from the wild canid population , leading us to consider these variants as derived alleles which were likely selected by humans to create large and muscled breeds ., The ACSL4 gene is associated with the traits of heavy muscling and “back fat thickness” in pigs , a phenotype that aptly describes the breeds carrying the mutation 57–61 ., We conclude that ACSL4 , potentially in concert with the upstream deletion in IRS4 , is needed to create the large bulky/muscled phenotype observed in the breeds reported here ( Figs 3 and 4 ) ., The 1 , 997 variants at locus 2 also define a large homozygous haplotype found in all large breeds , except the Scottish Deerhound ( S2 Fig ) ., The haplotype found in the large breeds is also observed in 24 of 66 small/medium breeds including the Boston Terrier , Boxer , French Bulldog , Irish Water Spaniel and Labrador Retriever ., Not unexpectedly , we observe the heterozygous state in 19 additional small breeds ., While WGS demonstrates that the haplotype is found in breeds of varying size , the fact that it is present in 18 of 19 breeds in the homozygous state suggest that it is necessary , but not sufficient , for large body mass ., Within this region , we detected missense changes in three genes: ARHGAP36 ( Rho GTPase Activating protein 36 ) , IGSF1 ( Immunoglobulin Superfamily Member 1 ) and FRMD7 ( FERM Domain Containing 7 ) ( S4 Table ) ., Since the IGSF1 gene is a strong candidate for body size 50–52 , we examined it further , noting three variants in the canine sequence ( S2 Fig ) ., The first is a single nucleotide change in the 3’UTR ( chrX . g . 102360204G>A; rs24856221 ) , but the distribution of the genotype in the dog population suggests that it was not associated with SBW ( S4 Table ) ., The second is a missense mutation in exon 12 ( chrX . g . 102364864T>G; rs852386368 ) that changes an aspartic acid to a glutamic acid ( ENSCAFP00000027740 . 3:p . Asp768Glu ) ., The codon is highly conserved in mammals ( Table 4 ) and is an excellent functional candidate with a likely high impact on protein function ( Polyphen score = 0 . 992 ) ., The third variant is an in-frame deletion ( chrX . g . 102369488-102369489insAAC; rs850984482 ) in exon 6 of the gene , which is in LD with the missense mutation ., The deletion removes one polar amino acid , asparagine , in the conserved immunoglobulin-like domain ( ENSCAFP00000027740 . 3:p . Asp376_Glu377insAsn ) and is also a strong functional candidate ., The two potentially functional IGSF1 mutations at locus 2 were considered further ., To determine the ancestral allele for each , we used Sanger sequencing to ascertain genotypes from a set of wild canids , including 24 grey wolves from geographically diverse areas , two red wolves , and two coyotes ., The two mutations ( missense SNP at exon 12 and deletion at exon 6 ) , while associated with large size in dogs , were never observed in the coyote , red wolves , or grey wolves , leading us to term these two large breed variants as “derived” alleles ., To determine the frequency of each candidate variant in domestic breeds , we used Sanger sequencing to analyze a large panel of 561 dogs encompassing 96 breeds ( S6 Table ) ., This panel included 10 additional large breeds and 36 more small/medium breeds ., We observe both the exon 6 and 12 variations of IGSF1 in the homozygous state in several large breeds of varying mass and skeletal size including the Bullmastiff , Great Dane , Great Pyrenees , Irish Wolfhound , Newfoundland , and Saint Bernard ., Heterozygous genotypes were also identified in six additional large breeds: Black Russian Terrier , Dogo Canario , Greater Swiss Mountain Dog , Mastino Abruzzese and Tibetan Mastiff ., As expected , based on this pattern , we never found the “large IGSF1 allele” for either variant in the five unrelated Sanger sequenced Scottish Deerhounds , which are included in our definition of large breeds ., We identified 17 medium and small breeds for which all sequenced dogs were homozygous for the derived allele at both IGSF1 variants , and 21 medium and small breeds that were heterozygous , recapitulating the pattern observed above ., Interestingly , among the medium and small breeds , we found muscled breeds such as Boston Terrier , Boxer , Bulldog , French Bulldog , Miniature Bull Terrier and Shar-pei ., The remaining 36 small/medium breeds with SBW ranging from 2 . 7 to 39 . 5 kg ( 6 . 1 to 87 lb ) , and corresponding to 48 . 7% of the medium/small breeds , were homozygous for the ancestral alleles for both variants ( S6 Table ) ., The “large alleles” were found in 95% of large breeds , and the genotypes appeared fixed ( homozygous for the “large alleles” ) in 76 . 2% of large breeds ., By comparison , 51 . 4% of medium/small dogs carry what we considered to be “large alleles” ( homozygous in 44 . 7% ) ., This argues that while IGSF1 likely plays a role in modulating weight variation in modern breeds , it is also , and more precisely , a contributor to the muscled phenotype in breeds spanning a range of body sizes ( Fig 4 ) ., In this study , we identified two loci on the X chromosome associated with SBW in domestic dog breeds , using a panel of 855 dogs selected to represent the full range of canine body size , which we genotyped on the Illumina Canine HD SNP array ., We showed that two large haplotypes at two loci were shared by the majority ( >90% ) of large breeds with SBW >41 kg ( 90 lb ) , for which derived alleles ( not present in wolves ) have been identified ., Fine mapping using whole genome sequencing data from 163 dogs revealed candidate variants in IRS4 and IGSF1 that are strongly associated with large breeds ., Interestingly , we also identified a phenotype of bulky or stocky build , which is also referred to as “heavily muscled , ” for which a third candidate gene , ACSL4 , and variant were associated ., The bulky haplotype was found post hoc and not detected by either our GWAS or any previously published GWAS , because no SNP on the canine HD SNP array is in LD with the variants ., These particular allelic distributions in the canine population highlight the strong impact of X chromosome genes in determining the weight and muscling of modern dog breeds ., Our previous studies identified alleles in the GHR , HMGA2 , SMAD2 , and STC2 genes as major contributors to SBW 47 ., When we included genotyping data from IGF1 and IGF1R , which we had identified previously as body size genes in dogs 41 , 42 , we showed that these six genes explain about 60% of body size variance in small/medium breeds , but <5% of variance in large breeds ., This highlights a now recurring theme in dog genetics that a small number of genes of large effect control many complex phenotypes , as opposed to many genes of small effect as is observed often in humans ., We used two different approaches to identify variants associated with large body size ., SNP chip data were used to identify large regions of LD ., However , this strategy does not detect rare variants that are not in LD ., WGS provides a complementary tool for these types of analyses ., Indeed , this allows detection of rare mutations that would otherwise go unnoticed ., In this study , the combination of dense SNP chip data ( Illumina 170k ) and WGS highlighted rare variants , such as the ACSL4 mutation , which are specific to a subset of large breeds , a result not found with SNP chip data alone ., This approach allowed us to define a new and very specific phenotype , the heavy muscling trait , which had not been previously described in dogs at a genetic level ., We found first that IRS4 is strongly associated with large body size in dogs ., The gene encodes a cytoplasmic protein that contains several potential tyrosine and serine/threonine phosphorylation sites ., IRS4 interacts with multiple growth factor receptors such as IGF1R , enhancing IGF1-stimulated cell growth 49 ., This gene is highly expressed in the hypothalamus which itself plays a primary role in regulation of body weight 56 ., It is also estrogen-regulated 62 , which may explain , in part , the established link between estrogen and body fat distribution 63 ., Moreover , a double “knock-out” mouse model ( bIrs2-/- . Irs4-/y ) developed severe obesity suggesting that IRS4 synergizes and complements IRS2 64 ., In humans , six SNPs in IRS4 have been identified that are associated with obesity , albeit in a cohort of patients with schizophrenia 56 ., In our study , we identified three genomic variations in IRS4 ., Neither the codon deletion nor one codon insertion in the exonic region of IRS4 appeared to be associated with disruptions in protein function ., However , in large bulky/muscled breeds , we also detected an associated 56 kb deletion located 150 kb upstream of the start codon of IRS4 ., This deletion contained several repeated elements , and may contain regulatory elements that affect the expression profile of IRS4 65 , 66 ., While no correlation was found between height and IRS4 in the human study 56 , in our canine study we observe a strong correlation between IRS4 and SBW that extends to include standard breed height ( SBH ) ( S4 Fig ) ., SBH is the height range assigned by the AKC for a given breed ., However , the addition of the SBH as co-variate in primary GWAS results in the loss of the locus 1 signal ., Interestingly , the reverse analysis confirms the strong association between SBW and both loci ., Indeed , the addition of SBW as a covariate for the SBH GWAS results in the loss of both signals on the X chromosome ( S4 Fig ) ., Overall this suggests that while both IRS4 and IGSF1 , the latter of which is the second candidate gene on X chromosome , are associated with variation in breed size , IRS4 is necessary , but not sufficient , for increasing size ., We also showed that the IGSF1 gene , positioned at a second locus on the X chromosome , is strongly associated with large dog breeds ., This gene encodes a plasma membrane glycoprotein and is involved in the thyroid hormone pathway 50 ., In large dog breeds , we identified two mutations , one single codon deletion and one missense mutation , both of which are located in a highly conserved immunoglobulin-like domain of IGSF1 protein ., In humans , mutations in the same IGSF1 protein domain are associated with the X-linked IGSF1 deficiency syndrome 50–52 , 67–69 ., Some patients show growth hormone ( GH ) deficiency during childhood , and 67% of male children are reportedly overweight while 21% are obese ( Review in 70 ) ., The general observation is supported by the fact that Igsf1-deficient male mice show diminished pituitary and serum thyroid-stimulating hormone ( TSH ) concentrations , reduced pituitary thyrotropin-releasing hormone ( TRH ) receptor expression , and increased body mass 50 ., Measuring these hormone levels in dogs , while difficult , may confirm the parallels between dogs and mice ., We also detected a strong association between IGSF1 and SBH ( S4 Fig ) ., Human studies used body mass index ( BMI ) as a measure of obesity given a particular height ., To date , 97 loci are associated with human BMI 71 ., It could be interesting to develop the same body mass index measure for dogs to better understand the results regarding IGSF1 , IRS4 , ACSL4 , IGF1 , IGF1R , HMAGA2 , GHR , SMAD2 , and STC2 ., This approach could explain why our study revealed that 50% of small/medium breed dogs have the “large alleles , ” mainly found in muscled breeds such as Boston Terrier or French Bulldog ( Fig 4 ) ., Interestingly , the IGSF1 locus also appears to be under selection in GWAS studies for other morphologic traits , such as brachycephalic ( e . g . bulldog , pug ) versus dolichocephalic ( e . g . afghan hound , collie ) skull shape 22 , 72 ( Fig 4 ) ., In humans , patients with microduplication of the IGSF1 locus present syndromic facial appearance 73 ., The varying phenotypes associated with IGSF1 illustrate the intermingling of genes and phenotypes regarding skeletal formation ., In addition to breed standard weight and heights , this study revealed a genetic association with a well-defined phenotype of bulkiness , due to heavy muscling and fat , which we found to be strongly associated with a highly conserved single nucleotide in the 5’ UTR in canine ACSL4 at locus 1 ( S3 Fig ) ., ACSL4 belongs to the long-chain acyl-CoA synthetase ( ACSL ) family and five genes have been identified in mammals ( ACSL1 , 3 , 4 , 5 , and 6 ) 74 , 75 ., ACSL4 binds specifically to longer chain polyunsaturated fatty acids ., While ACSL4 plays a role in many cellular processes 76–80 , increased ACSL4 expression in the liver likely promotes fatty acid uptake 53 ., The relationship between the gene and body shape in dogs fits well with this observation ., We did not observe the same relationship between ACSL4 and stocky dogs from small breeds , suggesting that the genetic variant found in large dogs is not relevant in the absence of genes that increase body size ., In the pig , mutations in ACSL4 are associated with a phenotype termed “back fat thickness ( BFT ) ” ., There are 75 common breeds of pigs ( http://www . thepigsite . com/ ) and large variation in adiposity between breeds has been described 58 ., Pig breeds with considerable back fat are used to study human obesity as well as obesity-related diseases , such as metabolic syndrome 81 ., Four QTLs on the porcine X chromosome were associated with the BFT , muscle mass , and intramuscular fat content 57 , 59 ., Post-mortem studies reveal that polymorphisms surrounding the ACSL4 gene are associated with BFT and muscle-associated traits in a pig breed-specific manner 57 as was observed in dog breeds within our study ., Specifically , the canine variant ( chrX . g . 82919525C>T ) was observed only in bulky dogs including , for instance , the Bullmastiff , Greater Swiss Mountain Dog , Newfoundland and Saint Bernard ., All of these breeds are well-muscled breeds compared to the leaner Great Dane , Borzoi which , for example , lack the variant ., The absence of the derived ACSL4 allele in more than 97% of breeds which meet the definition of medium/small , as well as in giant thin breeds led us to define the “bulky phenotype” in dogs characterized by the traits of heavy muscling and back fat thickness which , together , are observed in 54% of the large breeds ., We also notice an interesting correlation between the presence of the derived allele in some “large breeds” and their historic geographic distribution ( S5 Fig ) ., The “bulky allele” seems to have appeared in England-France ( Dogue de Bordeaux , English and Bullmastiff ) , become fixed in these breeds , and then spread through Europe ( Bernese Mountain Dog , Leonberg , Kuvasz ) ., Mediterranean and Eurasian breeds ( Cane Corso , Neapolitan Mastiff , Anatolian Shepherd ) do not have this allele , likely reflecting the recent geographic spread of the allele in Europe ., Finally , additional studies in pigs describe two mutations in the IRS4 gene , perhaps suggesting a second role for IRS4 as a contributor to BFT as well as general body size 57 , 82 ., In this study , we utilized WGS and GWAS to identify genes highly associated with large body size in dogs ., Modern dogs display a range of traits that have been easily mapped by taking advantage of the long LD observed in many breeds ., That same LD makes it problematic to go from associated marker to gene ., The availability of WGS represents a major advance for tackling this issue and , in this case , allowed us to disentangle the genetics of a complex trait on a relatively homogenous chromosome ., While a large number of genes of small effect seem to control body size in humans , in dogs a surprisingly small number of genes of large effect explain the range in size observed across breeds ., As dogs at the extremes of the body size continuum are studied , it will be interesting to note if genes previously identified from human studies are identified , or if an entirely new repertoire of genes are found which contribute to gigantism or miniaturization of breeds ., Studies in domestic dogs , therefore , provide a mechanism for understanding the genetics that underlies traits of interest in both human and domesticated animals ., Whole blood samples were collected into EDTA or ACD anticoagulant from AKC-registered dogs ., Genomic DNA was extracted using a standard phenol-chloroform extraction protocol 83 ., All procedures were reviewed and approved by the NHGRI Animal Care and Use Committee at the National Institutes of Health ., Standard breed weights and height were obtained from several sources: weights previously listed in Rimbault et al . 47 were used , although they were updated if weights specified by the AKC 84 were different ., If the AKC did not specify SBW and SBH , we used data from Atlas of Dog Breeds of the World 16 ., SBW and SBH ( male + female average ) were applied to all samples from the same breed and the values used in this study are listed in S1 and S2 Tables ., Analyses by sex did not change the results , thus we retained the genotypes as a single dataset ., Genotyping was performed using the Illumina 170K Canine HD SNP array containing approximately 170 , 000 SNPs distributed across the 38 canine autosomes and the X chromosome ., Genotypes were called using Illumina Genome Studio software ., In total , 855 dogs , 418 males and 437 females , were genotyped 21 ., Dogs belong to 88 different breeds ., Eighty-two breeds with nine to 11 dogs were genotyped and six large dog breeds with four to six dogs genotyped ., All samples had a call rate greater than 93% ( range: 93 . 57–99 . 98 , average: 99 . 84 ) ., SNPs with a minor allele frequency <1% or the presence of >5% missing genotypes were pruned , resulting in a final dataset of 150 , 895 SNPs that were used for the subsequent GWAS ., The GWAS was conducted using the software GEMMA v0 . 94 . 1 ( Genome-wide Efficient Mixed-Model Association ) 54 , 55 as a linear mixed-model software using a centered kinship matrix ., Pedigrees of dogs used in the study were verified to avoid inclusion of close relatives , i . e . none shared a common grandparent ., In the two regions of interest , pairwise r2 values were calculated using Pl

Introduction, Results, Discussion, Methods

Domestic dog breeds display significant diversity in both body mass and skeletal size , resulting from intensive selective pressure during the formation and maintenance of modern breeds ., While previous studies focused on the identification of alleles that contribute to small skeletal size , little is known about the underlying genetics controlling large size ., We first performed a genome-wide association study ( GWAS ) using the Illumina Canine HD 170 , 000 single nucleotide polymorphism ( SNP ) array which compared 165 large-breed dogs from 19 breeds ( defined as having a Standard Breed Weight ( SBW ) >41 kg 90 lb ) to 690 dogs from 69 small breeds ( SBW ≤41 kg ) ., We identified two loci on the canine X chromosome that were strongly associated with large body size at 82–84 megabases ( Mb ) and 101–104 Mb ., Analyses of whole genome sequencing ( WGS ) data from 163 dogs revealed two indels in the Insulin Receptor Substrate 4 ( IRS4 ) gene at 82 . 2 Mb and two additional mutations , one SNP and one deletion of a single codon , in Immunoglobulin Superfamily member 1 gene ( IGSF1 ) at 102 . 3 Mb ., IRS4 and IGSF1 are members of the GH/IGF1 and thyroid pathways whose roles include determination of body size ., We also found one highly associated SNP in the 5’UTR of Acyl-CoA Synthetase Long-chain family member 4 ( ACSL4 ) at 82 . 9 Mb , a gene which controls the traits of muscling and back fat thickness ., We show by analysis of sequencing data from 26 wolves and 959 dogs representing 102 domestic dog breeds that skeletal size and body mass in large dog breeds are strongly associated with variants within IRS4 , ACSL4 and IGSF1 .

Modern dog breeds display significant variation in body size and mass resulting from selective breeding practices ., A genome-wide association study ( GWAS ) of 170 , 000 SNPs genotyped on a panel of 855 dogs from 88 breeds revealed two loci on the canine X chromosome that were strongly associated with the large Standard Breed Weight ( SBW >41kg 90lb ) in domestic dog breeds ., Fine mapping of both loci using whole genome sequencing ( WGS ) data from 163 dogs highlighted three candidate genes , IRS4 and ACSL4 within the first locus ( 82–84 Mb ) , and IGSF1 at the second locus ( 101–104 Mb ) ., Associated variants were found in all three genes ., Of interest , the IRS4 gene is involved in the IGF-1/growth hormone pathway and IGSF1 is associated with human obesity ., ACSL4 is associated with muscling and back fat thickness in pigs , a phenotype we observe in “bulky” dog breeds ., This work identifies three new genes and associated variants that contribute to body mass in dogs , advancing our understanding of morphologic variability in domestic dog breeds .

genome-wide association studies, animal types, medicine and health sciences, variant genotypes, vertebrates, pets and companion animals, dogs, mammals, alleles, animals, genetic mapping, physiological parameters, genome analysis, zoology, sex chromosomes, chromosome biology, x chromosomes, genetic loci, cell biology, heredity, physiology, genetics, biology and life sciences, genomics, amniotes, computational biology, organisms, chromosomes, human genetics

journal.pcbi.1000464

Recognizing Sequences of Sequences

Many aspects of our sensory environment can be described as dynamic sequences ., For example , in the auditory domain , speech and music are sequences of sound-waves 1 , 2 , where speech can be described as a sequence of phonemes ., Similarly , in the visual domain , speaking generates sequences of facial cues with biological motion 3 , 4 ., These auditory and visual sequences have an important characteristic: the transitions between the elements are continuous; i . e . , it is often impossible to identify a temporal boundary between two consecutive elements ., For example , phonemes ( speech sounds ) in a syllable are not discrete entities that follow each other like beads on a string but rather show graded transitions to the next phoneme ., These transitions make artificial speech recognition notoriously difficult 5 ., Similarly , in the visual domain , when we observe someone speaking , it is extremely difficult to determine exactly where the movements related to a phoneme start or finish ., These dynamic sequences , with brief transitions periods between elements , are an inherent part of our environment , because sensory input is often generated by the fluent and continuous movements of other people , or indeed oneself ., Dynamic sequences are generated on various time-scales ., For example , in speech , formants form phonemes and phonemes form syllables ., Sequences , which exist at different time-scales , are often structured hierarchically , where sequence elements on one time-scale constrain the expression of sequences on a finer time-scale; e . g . a syllable comprises a specific sequence of phonemes ., This functional hierarchy of time-scales may be reflected in the hierarchical , anatomical organisation of the brain 6 ., For example , in avian brains , there is anatomical and functional evidence that birdsong is generated and perceived by a hierarchical system , where low levels represent transient acoustic details and high levels encode song structure at slower time-scales 7 , 8 ., An equivalent temporal hierarchy might also exist in the human brain for representing auditory information , such as speech 1 , 9–12 ., Here we ask the following question: How does the brain recognize the dynamic and ambiguous causes of noisy sensory input ?, Based on experimental and theoretical evidence 13–18 we assume the brain is a recognition system that uses an internal model of its environment ., The structure of this model is critical: On one hand , the form of the model must capture the essential architecture of the process generating sensory data ., On the other hand , it must also support robust inference ., We propose that a candidate that fulfils both criteria is a model based on a hierarchy of stable heteroclinic channels ( SHCs ) ., SHCs have been introduced recently as a model of neuronal dynamics per se 19 ., Here , we use SHCs as the basis of neuronal recognition , using an established Bayesian scheme for modelling perception 20 ., This brings together two recent developments in computational approaches to perception: Namely , winnerless competition in stable heteroclinic channels and the hypothesis that the brain performs Bayesian inference ., This is important because it connects a dynamic systems perspective on neuronal dynamics 19 , 21 , 22 with the large body of work on the brain as an inference machine 13–18 ., To demonstrate this we generate artificial speech input ( sequences of syllables ) and describe a system that can recognize these syllables , online from incoming sound waves ., We show that the resulting recognition dynamics display functional characteristics that are reminiscent of psychophysical and neuronal responses ., SHCs are attractors formed by artificial neuronal networks , which prescribe sequences of transient dynamics 22–25 ., The key aspect of these dynamical systems is that their equations of motion describe a manifold with a series of saddle points ., At each saddle point , trajectories are attracted from nearly all directions but are expelled in the direction of another saddle point ., If the saddle points are linked up to form a chain , the neuronal state follows a trajectory that passes through all these points , thereby forming a sequence ., These sequences are exhibited robustly , even in the presence of high levels of noise ., In addition , the dynamics of the SHCs are itinerant due to dynamical instability in the equations of motion and noise on the states ., This noise also induces a variation in the exact times that sequence elements are visited ., This can be exploited during recognition , where the SHC places prior constraints on the sequence that elements ( repelling fixed-points ) are visited but does not constrain the exact timing of these visits ., The combination of these two features , robustness of sequence order but flexibility in sequence timing , makes the SHC a good candidate for the neuronal encoding of trajectories 19 , 26 ., Rabinovich et al . have used SHCs to explain how spatiotemporal neuronal dynamics observed in odour perception , or motor control of a marine mollusc , can be expressed in terms of a dynamic system 22 , 27 ., Varona et al . used Lotka-Volterra-type dynamics to model a network of six neurons in a marine mollusc 27: With particular lateral inhibition between pairs of neurons and input to each neuron , the network displayed sequences of activity ., Following a specific order , each neuron became active for a short time and became inactive again , while the next neuron became active , and so on ., Stable heteroclinic channels rest on a particular form of attractor manifold that supports itinerant dynamics ., This itinerancy can result from deterministic chaos in the absence of noise , which implies the presence of heteroclinic cycles ., When noise is added , itinerancy can be assured , even if the original system has stable fixed-points ., However , our motivation for considering stochastic differential equations is to construct a probabilistic model , where assumptions about the distribution of noise provide a formal generative model of sensory dynamics ., As reviewed in 22 , Lotka-Volterra dynamics can be derived from simple neural mass models of mean membrane potential and mean firing rate 21 ., Here , we use a different neural mass model , where the state-vector x can take positive or negative values: ( 1 ) where the motion of a hidden-state vector ( e . g . , mean membrane potentials ) x is a nonlinear function of itself with scalar parameters , , and a connectivity matrix ., The hidden state-vector enters a nonlinear function S to generate outcomes ( e . g . , neuronal firing rates ), y . Each element determines the strength of lateral inhibition from state j to i ., Both the state and observation equations above include additive normally distributed noise vectors w and z ., When choosing specific parameter values ( see below ) , the states display stereotyped sequences of activity 28 ., Rabinovich et al . 19 termed these dynamics ‘stable heteroclinic channels’ ( SHCs ) ., If the channel forms a ring , once a state is attracted to a saddle point , it will remain in the SHC ., SHCs represent a form of itinerant dynamics 26 , 29 , 30 and may represent a substrate for neuronal computations 31 ., Remarkably , the formation of SHCs seems to depend largely on the lateral inhibition matrix and not on the type of neuronal model; see Ivanchenko et al . 32 for an example using a complex two-compartment spiking neuron model ., In this paper , we propose to use SHCs not as a model for neuronal dynamics per se but as a generative model of how sensory input is generated ., This means that we interpret x as hidden states in the environment , which generate sensory input, y . The neuronal response to sampling sensory input y are described by recognition dynamics , which decode or deconvolve the causes x from that input ., These recognition dynamics are described below ., This re-interpretation of Eq ., 1 is easy to motivate: sensory input is usually generated by our own body and other organisms ., This means input is often generated by neuronal dynamics of the sort described in Eq ., 1 . A SHC can generate repetitive , stereotyped sequences ., For example , in a system with four saddle points , an SHC forces trajectories through the saddle points in a sequence , e . g . ‘1-2-3-4-1-2-3-4-1…’ ., In contrast , a SHC cannot generate ‘1-2-3-4-3-4-2-1…’ , because the sequence is not repetitive ., However , to model sensory input , for example speech , one must be able to recombine basic sequence-elements like phonemes in ever-changing sequences ., One solution would be to represent each possible sequence of phonemes ( e . g . each syllable ) with a specific SHC ., A more plausible and parsimonious solution is to construct a hierarchy of SHCs , which can encode sequences generated by SHCs whose attractor topology ( e . g . the channels linking the saddle points ) is changed by a supraordinate SHC ., This can be achieved by making the connectivity matrix at a subordinate level a function of the output states of the supra-ordinate level ., This enables the hierarchy to generate sequences of sequences to any hierarchical depth required ., Following a recent account of how macroscopic cortical anatomy might relate to time-scales in our environment 6 , we can construct a hierarchy by setting the rate constant of the j-th level to a rate that is slower than its subordinate level , ., As a result , the states of subordinate levels change faster than the states of the level above ., This means the control parameters at any level change more slowly than its states , ; because the slow change in the attractor manifold is controlled by the supraordinate states: ( 2 ) where the superscript indexes level j ( level 1 being the lowest level ) , are ‘hidden states’ , and are outputs to the subordinate level , which we will call ‘causal states’ ., As before , at the first level , is the sensory stream ., In this paper , we consider hierarchies with relative time-scales of around four ., This means that the time spent in the vicinity of a saddle point at a supraordinate level is long enough for the subordinate level to go through several saddle points ., As before , all levels are subject to noise on the motion of the hidden states and the causal states ., At the highest level , the control parameters , are constant over time ., At all other levels , the causal states of the supraordinate level , , enter the subordinate level by changing the control parameters , the connectivity matrix : ( 3 ) Here , is a linear mixture of ‘template’ control matrices , weighted by the causal states at level ., Each of these templates is chosen to generate a SHC ., Below , we will show examples of how these templates can be constructed to generate various sequential phenomena ., The key point about this construction is that states from the supraordinate level select which template controls the dynamics of the lower level ., By induction , the states at each level follow a SHC because the states at the supraordinate level follow a SHC ., This means only one state is active at any time and only one template is selected for the lower level ., An exception to this is the transition from one state to another , which leads to a transient superposition of two SHC-inducing templates ( see below ) ., Effectively , the transition transient at a specific level gives rise to brief spells of non-SHC dynamics at the subordinate levels ( see results ) ., These transition periods are characterized by dissipative dynamics , due to the largely inhibitory connectivity matrices , inhibition controlled by parameter ( Eq . 2 ) and the saturating nonlinearity S . In summary , a hierarchy of SHCs generates the sensory stream at the lowest ( fastest ) level , which forms a sequence of sequences expressed in terms of first-level states ., In these models , the lower level follows a SHC , i . e . the states follow an itinerant trajectory through a sequence of saddle points ., This SHC will change whenever the supraordinate level , which follows itself a SHC , moves from one saddle point to another ., Effectively , we have constructed a system that can generate a stable pattern of transients like an oscillator; however , as shown below , the pattern can have deep or hierarchical structure ., Next , we describe how the causes can be recognized or deconvolved from sensory input, y . We have described how SHCs can , in principle , generate sequences of sequences that , we assume , are observed by an agent as its input, y . To recognise the causes of the sensory stream the agent must infer the hidden states online , i . e . the system does not look into the future but recognizes the current states and of the environment , at all levels of the hierarchy , by the fusion of current sensory input and internal dynamics elicited by past input ., An online recognition scheme can be derived from the ‘free-energy principle’ , which states that an agent will minimize its surprise about its sensory input , under a model it entertains about the environment; or , equivalently maximise the evidence for that model 18 ., This requires the agent to have a dynamic model , which relates environmental states to sensory input ., In this context , recognition is the Bayesian inversion of a generative model ., This inversion corresponds to mapping sensory input to the posterior or conditional distribution of hidden states ., In general , Bayesian accounts of perception rest on a generative model ., Given such a model , one can use the ensuing recognition schemes in artificial perception and furthermore compare simulated recognition dynamics ( in response to sensory input ) , with evoked responses in the brain ., The generative model in this paper is dynamical and based on the nonlinear equations 1 and 2 . More precisely , these stochastic differential equations play the role of empirical priors on the dynamics of hidden states causing sensory data ., In the following , we review briefly , the Bayesian model inversion described in 20 for stochastic , hierarchical systems and apply it , in the next section , to hierarchical SHCs ., Given some sensory data vector y , the general inference problem is to compute the model evidence or marginal likelihood of y , given a model m: ( 4 ) where the generative model is defined in terms of a likelihood and prior on hidden states ., In Equation 4 , the state vector subsumes the hidden and causal states at all levels of a hierarchy ( Eq . 2 ) ., The model evidence can be estimated by converting this difficult integration problem ( Eq . 4 ) into an easier optimization problem by optimising a free-energy bound on the log-evidence 33 ., This bound is constructed using Jensens inequality and is a function of an arbitrary recognition density , : ( 5 ) The free-energy comprises an energy term and an entropy term and is defined uniquely , given a generative model ., The free-energy is an upper bound on the surprise or negative log-evidence , because the Kullback-Leibler divergence , between the recognition and conditional density , is always positive ., Minimising the free-energy minimises the divergence , rendering the recognition density an approximate conditional density ., When using this approach , one usually employs a parameterized fixed-form recognition density , 20 ., Inference corresponds to optimising the free-energy with respect to the sufficient statistics , of the recognition density: ( 6 ) The optimal statistics are sufficient to describe the approximate posterior density; i . e . the agents belief about ( or representation of ) the trajectory of the hidden and causal states ., We refer the interested reader to Friston et al . 34 for technical details about this variational Bayesian treatment of dynamical systems ., Intuitively , this scheme can be thought of as augmented gradient descent on a free-energy bound on the models log-evidence ., Critically , it outperforms conventional Bayesian filtering ( e . g . , Extended Kalman filtering ) and eschews the computation of probability transition matrices ., This means it can be implemented in a simple and neuronally plausible fashion 20 ., In short , this recognition scheme operates online and recognizes current states of the environment by combining current sensory input with internal recognition dynamics , elicited by past input ., A recognition system that minimizes its free-energy efficiently will come to represent the environmental dynamics in terms of the sufficient statistics of recognition density; e . g . the conditional expectations and variances of ., We assume that the conditional moments are encoded by neuronal activity; i . e . , Equation 6 prescribes neuronal recognition dynamics ., These dynamics implement Bayesian inversion of the generative model , under the approximations entailed by the form of the recognition density ., Neuronally , Equation 6 can be implemented using a message passing scheme , which , in the context of hierarchical models , involves passing prediction errors up and passing predictions down , from one level to the next ., These prediction errors are the difference between the causal states ( Equation 2 ) ; ( 7 ) at any level j , and their prediction from the level above , evaluated at the conditional expectations 18 , 35 ., In addition , there are prediction errors that mediate dynamical priors on the motion of hidden states within each level ( Equation 2 ) ; ( 8 ) This means that neuronal populations encode two types of dynamics: the conditional expectations of states of the world and the prediction errors ., The dynamics of the first are given by Equation 6 , which can be formulated as a function of prediction error ., These dynamics effectively suppress or explain away prediction error; see 34 for details ., This inversion scheme is a generic recognition process that receives dynamic sensory input and can , given an appropriate generative model , rapidly identify and track environmental states that are generating current input ., More precisely , the recognition dynamics resemble the environmental ( hidden ) states they track ( to which they are indirectly coupled ) , but differ from the latter because they are driven by a gradient descent on free-energy; Eq ., 6 ( i . e . minimize prediction errors: Eqs . 7 and 8 ) ., This is important , because we want to use SHCs as a generative model , not as a model of neuronal encoding per se ., This means that the neuronal dynamics will only recapitulate the dynamics entitled by SHCs in the environment , if the recognition scheme can suppress prediction errors efficiently in the face of sensory noise and potential beliefs about the world ., We are now in a position to formulate hierarchies of SHCs as generative models , use them to generate sensory input and simulate recognition of the causal states generating that input ., In terms of low-level speech processing , this means that any given phoneme will predict the next phoneme ., At the same time , as phonemes are recognized , there is also a prediction about which syllable is the most likely context for generating these phonemes ., This prediction arises due to the learnt regularities in speech ., In turn , the most likely syllable predicts the next phoneme ., This means that speech recognition can be described as a dynamic process , on multiple time-scales , with recurrently evolving representations and predictions , all driven by the sensory input ., In the auditory system , higher cortical levels appear to represent features that are expressed at slower temporal scales 36 ., Wang et al . 37 present evidence from single-neuron recordings that there is a ‘slowing down’ of representational trajectories from human auditory sensory thalamus ( a ‘relay’ to the primary auditory cortex ) , the medial geniculate body ( MGB ) to primary auditory cortex ( AI ) ., In humans , it has been found that the sensory thalamus responds preferentially to faster temporal modulations of sensory signals , whereas primary cortex prefers slower modulations 10 ., These findings indicate that neuronal populations , at lower levels of the auditory system ( e . g . MGB ) , represent faster environmental trajectories than higher levels ( e . g . , A1 ) ., Specifically , the , MGB responds preferentially to temporal modulations of ∼20 Hz ( ∼50 ms ) , whereas AI prefers modulations at ∼6 Hz ( ∼150 ms ) 10 ., Such a temporal hierarchy would be optimal for speech recognition , in which information over longer time-scales provides predictions for processing at shorter time scales ., In accord with this conjecture , optimal encoding of fast ( rapidly modulated ) dynamics by top-down predictions has been found to be critical for communication 1 , 12 , 38 ., We model this ‘slowing down’ with a hierarchical generative model based on SHCs ., This model generates sequences of syllables , where each syllable is a sequence of phonemes ., Phonemes are the smallest speech sounds that distinguishes meaning and a syllable is a unit of organization for a sequence of phonemes ., Each phoneme prescribes a sequence of sound-wave modulations which correspond to sensory data ., We generated data in this fashion and simulated online recognition ( see Figure 1 ) ., By recognizing speech-like phoneme-sequences , we provide a proof-of-principle that a hierarchical system can use sensory streams to infer sequences ., This not only models the slowing down of representations in the auditory system 10 , 12 , 37 , 38 , but may point to computational approaches to speech recognition ., In summary , the recognition dynamics following Equation 6 are coupled to a generative model based on SHCs via sensory input ., The systems generating and recognising states in Fig . 1 are both dynamic systems , where a non-autonomous recognition system is coupled to an autonomous system generating speech ., All our simulations used hierarchies with two levels ( Figure 2 ) ., The first ( phonemic ) level produces a sequence of phonemes , and the second ( syllabic ) level encodes sequences of syllables ., We used Equation 2 to produce phoneme sequences , where the generating parameters are listed in Table 3 . The template matrices ( Equation 3 ) were produced in the following way: We first specified the sequence each template should induce; e . g . , sequence 1-2-3 for three neuronal populations ., We then set elements on the main diagonal to 1 , the elements ( 2 , 1 ) , ( 3 , 2 ) , ( 1 , 3 ) to value 0 . 5 , and all other elements to 5 28 ., More generally for sequence ( 9 ) Note that SHC hierarchies can be used to create a variety of different behaviours , using different connectivity matrices ., Here we explore only a subset of possible sequential dynamics ., When generating sensory data y , we added noise and to both the hidden and causal states ., At the first and second levels , this was normally distributed zero-mean noise with log-precisions of ten and sixteen , respectively ., These noise levels were chosen to introduce noisy dynamics but not to the extent that the recognition became difficult to visualise ., We repeated all the simulations reported below with higher noise levels and found that the findings remained qualitatively the same ( results not shown ) ., Synthetic stimuli were generated by taking a linear mixture of sound waves extracted from sound files , in which a single speaker pronounced each of four vowel-phonemes: a , e , i , o ., These extracts W were sampled at 22050 Hz and about 14 ms long ., The mixture was weighted by the causal states of the phonemic level; ., This resulted in a concatenated sound wave file w ., When this sound file is played , one perceives a sequence of vowels with smooth , overlapping transitions ( audio file S1 ) ., These transitions are driven by the SHCs guiding the expression of the phonemes and syllables at both levels of the generative hierarchy ., For computational simplicity , we circumvented a detailed generative model of the acoustic level ., For simulated recognition , the acoustic input ( the sound wave ) was transformed to phonemic input by inverting the linear mixing described above every seven ms of simulated time ( one time bin ) ., This means that our recognition scheme at the acoustic level assumes forward processing only ( Fig . 1 ) ., However , in principle , given an appropriate generative model 39 , 40 , one could invert a full acoustic model , using forward and backward message passing between the acoustic and phonemic levels ., To create synthetic stimuli we generated syllable sequences consisting of four phonemes or states; a , e , i , and o , over 11 . 25 seconds ( 800 time points ) , using a two-level SHC model ( Fig . 2 ) ., To simulate word-like stimuli , we imposed silence at the beginning and the end by windowing the phoneme sequence ( Fig . 3A , top left ) ., At the syllabic level , we used three syllables or states to form the second-level sequence ( 1–2–3 ) ( 2 ) ; where the numbers denote the sequence and the superscript indicates the sequence level ., The three causal states of the syllabic level entered the phonemic level as control parameters to induce their template matrices as in Equation 3 ., This means that each of the three syllable states at the second level causes a phoneme sequence at the first: , , and , see Fig . 2 and listen to the audio file S1 ., In Fig . 3A we show the causal and hidden states , at both levels , generated by this model ., The remaining parameters , for both levels , are listed in Table 3 ., Note that the rate constant of the syllabic level is four times slower than at the phonemic level ., As expected , the phoneme sequence at the first level changes as a function of the active syllable at the second level ., The transients caused by transitions between syllables manifest at the first level as temporary changes in the amplitude or duration of the active phoneme ., We then simulated recognition of these sequences ., Fig . 3B shows that our recognition model successfully tracks the true states at both levels ., Note the recognition dynamics rapidly ‘lock onto’ the causal states from the onset of the first phoneme of the first syllable ( time point 50 ) ., Interestingly , the system did not recognize the first syllable ( true: syllable 3 ( red line ) , recognized: syllable 2 ( green line ) between time points 50 to 80 ( see red arrow in Fig . 3B ) , but corrected itself fairly quickly , when the sensory stream indicated a new phoneme that could only be explained by the third syllable ., This initial transient at the syllabic level shows that recognition dynamics can show small but revealing deviations from the true state dynamics ., In principle , these deviations could be used to test whether the real auditory system uses a recognition algorithm similar to the one proposed; in particular , the simulated recognition dynamics could be used to explain empirical neurophysiological responses ., What happens if the stimuli deviate from learned expectations ( e . g . violation of phonotactic rules ) ?, In other words , what happens if we presented known phonemes that form unknown syllables ?, This question is interesting for two reasons ., First , our artificial recognition scheme should do what we expect real brains to do when listening to a foreign language: they should be able to recognize the phonemes but should not derive high-order ‘meaning’ from them; i . e . should not recognize any syllable ., Secondly , there are well-characterised brain responses to phonotactic violations , e . g . 41–43 ., These are usually event-related responses that contain specific waveform components late in peristimulus time , such as the N400 ., The N400 is an event-related potential ( ERP ) component typically elicited by unexpected linguistic stimuli ., It is characterized as a negative deflection ( topologically distributed over central-parietal sites on the scalp ) , peaking approximately 400 ms after the presentation of an unexpected stimulus ., To model phonotactic violations , we generated data with the two-level model presented above ., However , we used syllables , i . e . sequences of phonemes , that the recognition scheme was not informed about and consequently could not recognise ( it has three syllables in its repertoire: , , and ) ., Thus the recognition scheme knows all four phonemes but is unable to predict the sequences heard ., Fig . 4A shows that the recognition system cannot track the syllables; the recognized syllables are very different from the true syllable dynamics ., At the phonemic level , the prediction error deviates from zero whenever a new ( unexpected ) phoneme is encountered ( Fig . 4B ) ., The prediction error at the syllabic level is sometimes spike-like and can reach high amplitudes , relative to the typical amplitudes of the true states ( see Fig . 4A and B ) ., This means that the prediction error signals violation of phonotactic rules ., In Fig . 4C , we zoom in onto time points 440 to 470 to show how the prediction error evolves when evidence of a phonotactic violation emerges: At the phoneme level , prediction error builds up because an unexpected phoneme appears ., After time point 450 , the prediction error grows quickly , up to the point that the system resolves the prediction error ., This is done by ‘switching’ to a new syllable , which can explain the transition to the emerging phoneme ., The switching creates a large amplitude prediction error at time point 460 ., In other words , in face of emerging evidence that its current representation of syllables and phonemes cannot explain sensory input , the system switches rapidly to a new syllable representation , giving rise to a new prediction error ., It may be that these prediction errors are related to electrophysiological responses to violations of phonotactic rules , 44 , 45 ., This is because the largest contributors to non-invasive electromagnetic signals are thought to be superficial pyramidal cells ., In biological implementations of the recognition scheme used here 20 , these cells encode prediction error ., In summary , these simulations show that a recognition system cannot represent trajectories or sequences that are not part of its generative model ., In these circumstances , recognition experiences intermittent high-amplitude prediction errors because the internal predictions do not match the sensory input ., There is a clear formal analogy between the expression of prediction error in these simulations and mismatch or prediction violation responses observed empirically ., The literature that examines event-related brain potentials ( ERPs ) and novelty processing “reveals that the orienting response engendered by deviant or unexpected events consists of a characteristic ERP pattern , comprised sequentially of the mismatch negativity ( MMN ) and the novelty P3 or P3a” 46 ., Human speech recognition is robust to the speed of speech 47 , 48 ., How do our brains recognize speech at different rates ?, There are two possible mechanisms in our model that can deal with ‘speaker speed’ parameters online ., First , one could make the rate constants and free parameters and optimise them during inversion ., Adjusting to different speaker parameters is probably an essential faculty , because people speak at different speeds 49 ., The second mechanism is that the recognition itself might be robust to deviations from the expected rate of phonemic transitions; i . e . , even though the recognition uses the rate parameters appropriate for much slower speech , it still can recognize fast speech ., This might explain why human listeners can understand speech at rates that they have never experienced previously 47 ., In the following , we show that our scheme has this robustness ., To simulate speed differences we used the same two-level model as in the simulations above with for the generation of phonemes , but with for recognition so that the st

Introduction, Model, Results, Discussion

The brains decoding of fast sensory streams is currently impossible to emulate , even approximately , with artificial agents ., For example , robust speech recognition is relatively easy for humans but exceptionally difficult for artificial speech recognition systems ., In this paper , we propose that recognition can be simplified with an internal model of how sensory input is generated , when formulated in a Bayesian framework ., We show that a plausible candidate for an internal or generative model is a hierarchy of ‘stable heteroclinic channels’ ., This model describes continuous dynamics in the environment as a hierarchy of sequences , where slower sequences cause faster sequences ., Under this model , online recognition corresponds to the dynamic decoding of causal sequences , giving a representation of the environment with predictive power on several timescales ., We illustrate the ensuing decoding or recognition scheme using synthetic sequences of syllables , where syllables are sequences of phonemes and phonemes are sequences of sound-wave modulations ., By presenting anomalous stimuli , we find that the resulting recognition dynamics disclose inference at multiple time scales and are reminiscent of neuronal dynamics seen in the real brain .

Despite tremendous advances in neuroscience , we cannot yet build machines that recognize the world as effortlessly as we do ., One reason might be that there are computational approaches to recognition that have not yet been exploited ., Here , we demonstrate that the ability to recognize temporal sequences might play an important part ., We show that an artificial decoding device can extract natural speech sounds from sound waves if speech is generated as dynamic and transient sequences of sequences ., In principle , this means that artificial recognition can be implemented robustly and online using dynamic systems theory and Bayesian inference .

neuroscience/theoretical neuroscience, neuroscience/sensory systems, computational biology/computational neuroscience

journal.pcbi.1004737

Muscle Synergies Heavily Influence the Neural Control of Arm Endpoint Stiffness and Energy Consumption

Limb impedance control by the central nervous system ( CNS ) has been a subject of much study and debate over the past three decades ., Numerous experiments and theoretical analyses have studied the biomechanical and neuromuscular capabilities of the CNS to regulate the impedance of a limb ( e . g . , 1–22 ) ., The preferred paradigm of many studies is to analyze the stiffness the human arm can produce at its endpoint ( i . e . , the hand ) in reaching-like postures in a horizontal plane in front of a seated subject ., One set of experimental findings is that , after some training , the CNS can regulate to varying degrees the orientation and eccentricity of arm stiffness ellipses to perform a task more reliably and efficiently than before training 1 , 4 , 5 , 10 ., Another set of experiments concludes that the CNS cannot arbitrarily regulate endpoint stiffness , and that it is only able to rotate the orientation of the stiffness ellipsoid around 30° 15 , 20 , 21 ., Here we focus on reconciling some of these conflicting results by using novel computational analyses of tendon-driven systems to establish the neuromechanical capabilities of biological limbs in the context of muscle synergies ., The existence and interpretation of muscle synergies is controversial and has received much attention in the recent literature 23–29 ., Synergies—defined as the correlated activation of multiple muscles by using a small number of coordination patterns—are theoretically one way to simplify the control of movement in the highly redundant musculature of vertebrates ., They have also been observed by EMG measurements during reaching movements with the arm 19 ., Here we explore the restrictions synergies could impose on the ability of the CNS to synthesize arm endpoint stiffnesses with differing characteristics ., There is extensive literature on the analysis and synthesis of endpoint stiffness in robotic limbs 7 , 30–33 ., The theoretical contributions and conclusions of these robotics studies are independent of the mechanisms and limitations of sensorimotor control by the CNS , and hence form a good theoretical foundation to design and interpret experiments to study the neuromechanical capabilities of biological limbs both in the presence and absence of synergies ., In 34 such an approach was used to compare theoretical predictions against experimental findings by recording from a few finger muscles ., In this study , we investigate the effects of muscle synergies on endpoint stiffness synthesis and energy consumption ( Fig 1 ) ., To this end , we apply principles of robotics in a novel computational formulation for tendon-driven systems that allows us to easily and efficiently analyze the range over which the stiffness of the endpoint of the limb can be modified ., More specifically , we are referring to the magnitude of endpoint stiffness in a variety of directions which can be mathematically approximated by a stiffness ellipse ., From an engineering perspective , we can call this the range of ‘stiffness realizations’ because each of them is an instance of the neuromechanical capabilities of the limb ., By studying stiffness realizations in the presence and absence of muscle synergies throughout the workspace , we find that synergies drastically decrease the ability of the CNS to synthesize an arbitrary stiffness ellipse ., Importantly , our work takes on the approach that we are interested in finding the families of feasible endpoint stiffness realizations throughout the workspace of the limb ., That is , how much can the nervous system control the size , shape ( i . e . , eccentricity ) and orientation of endpoint stiffness ellipses for all limb postures ?, Due to muscle redundancy , there may be multiple ways to achieve any one possible stiffness ellipse ., That set of multiple neural commands that can achieve a given realization is its ‘feasible activation set’ 35 , 36 ., We can therefore optimize over that set to find the muscle activation pattern that produces the desired endpoint stiffness while minimizing energy consumption ., The question is , then , how do muscle synergies compromise—or even annihilate—the ability of the nervous system to control the properties of endpoint stiffness and minimize energy consumption ?, As mentioned in the Discussion , this neuromechanical approach emphasizes the feasibility of neuromechanical actions and allows us to consider several potential confounds when comparing across studies ., Such studies may include examination of the extent , efficacy and nature of training , the influence of limb postures on task goals specified by the experimenter , and the implicit neural strategies specified by the CNS with regard to stiffness regulation in health and disease ., Our results also allow us to discuss how learning , experimental design , and neural strategies affect our ability to tune endpoint stiffness ., We use a simplified planar arm model with 6 muscles similar to those that have been used in other theoretical and computational studies 3 , 8 , 12 , 13 , 37 ( Fig 2 ) ., In those studies as well as this study—as described below—the spatial distribution of the stiffness of the endpoint of the limb , Kend , is a matrix that is calculated as a function of individual musculotendon stiffnesses , which are the elements of the matrix Kmuscles ., The individual musculotendon stiffness generated by each muscle is represented by a numerical variable that is , in effect , a lumped parameter model introduced by Hogan and Mussa-Ivaldi 8 , 13 that combines the active and passive components of muscle and the passive components of tendon ., This approximation remains commonplace and valid in the computational literature whose goal is not to simulate the physiology of musculotendon stiffness , but rather use a mechanical analogue of musculotendons to allow the study of the feasible mechanical behavior of the limb ., This lumped parameter approach is accepted in the computational literature to replicate the fact that musculotendons have stiffness , and that stiffness can be modulated by the individual neural commands to the muscles of a limb , the activation vector a → ., The reader is referred to the literature for details 3 , 8 , 12 , 13 , 37 , but a brief description is presented below ., We use workspace constraints identical to those used in 8 to produce the workspace of the limb ( i . e . , the locations that are reachable by the endpoint ) , also shown in Fig 2 . As is common , we use singular value decomposition ( SVD ) to transform the endpoint stiffness matrix ( Kend ) to an ellipse that represents the characteristics of this matrix—shape ( can also be termed eccentricity ) —measured by the matrix condition number , and orientation of the major axis with respect to the x-axis ., We begin our formulation with the endpoint stiffness matrix , Kend , as a function of muscle active stiffnesses , Kmuscles ., As a point of clarification , we focus on muscle active stiffness that results from the feedforward activation of the muscle such as during force production or co-contraction ., We do not include passive stiffness , which normally refers to the inherent material properties of the limb or muscles with no muscle activation ., This could arise , for example , from tendon properties ., Similarly , as mentioned in the Discussion , we do not include the time-delayed stiffness resulting from reflexes , often called reflexive stiffness ., In our formulation , Kend relates the vector of differential endpoint displacements to differential endpoint forces:, ∂ F → = K e n d ∂ x → ( 1 ), where ∂ F → is the endpoint force vector resulting from a displacement vector ∂ x → ., The joint stiffness matrix , Kjoint , relates the vector of differential joint angle displacements to differential joint torques:, ∂ τ → = K j o i n t ∂ θ → ( 2 ), where ∂ τ → is the joint torque vector resulting from a joint angle displacement vector ∂ θ → ., The endpoint stiffness matrix is dependent on the joint stiffness matrix as well as the manipulator Jacobian J ( which is posture dependent: a vector of joint angles θ → uniquely defines the posture ) :, x → ˙ = J ( θ → ) θ → ˙ ( 3 ), where x → ˙ denotes the endpoint velocity vector and θ → ˙ denotes the joint angle velocity vector ., The endpoint stiffness matrix , in the absence of an external tip force , is given by 8:, K e n d = J - T K j o i n t J - 1 ( 4 ), Furthermore , the joint stiffness matrix is given by 8:, K j o i n t = R K m u s c l e s R T ( 5 ), where Kmuscles is the diagonal matrix of muscle stiffnesses and R is the moment arm matrix relating joint angle changes to tendon displacements , ∂ s →:, ∂ s → = R ∂ θ → ( 6 ) Combining Eqs 4 and 5 , we obtain the relationship of muscle stiffness to endpoint stiffness:, K e n d = J - T R K m u s c l e s R T J - 1 ( 7 ) This is equivalent to other formulations , such as in 34 ., The diagonal elements of Kmuscles are assumed to be linearly related to their corresponding muscle forces 38:, K m u s c l e s = α × d i a g ( F → m u s c l e s ) ( 8 ), For simplicity in this study , we assume the scaling factor α is equal to one ., We can define a diagonal matrix of maximal muscle forces , Fmax , so that we can calculate F → m u s c l e s using the muscle activation vector a →:, F → m u s c l e s = F m a x a → ( 9 ) The entries of a → are inside the interval 0 , 1 since muscle force can only be positive ( this constraint can also be expressed as the requirement that the activation vector lies in the positive orthant of the unit hypercube in activation space ) ., We assume Fmax to be the identity matrix for simplicity in this study ., Using Eqs 7 , 8 and 9 and reformulating the endpoint stiffness matrix , the moment arm matrix , and the Jacobian , we can make the endpoint stiffness K ˜ e n d a vector that is a linear function of the muscle activations ., K ˜ e n d = J ˜ - T R ˜ F m a x a → ( 10 ) We show these reformulations in Fig 3 . ( • ) denotes element-by-element multiplication , and Ri is the ith row of R . The Jacobian reformulation is specific to the 2-link planar arm model , but similar expressions can be formulated for Jacobians of higher dimensions ., The endpoint stiffness and the moment arm matrices have been previously reformulated in this way 33 ., And 34 speaks of the equations defining iso-stiffness planes ., But to the best of our knowledge , no study has yet reformulated the Jacobian in this way to allow for the simple set of linear equations found in Eq 10 relating muscle activations to endpoint stiffness ., Each realization of a given endpoint stiffness matrix—and its associated ellipse—is produced by a given neural command , a → , as shown in Eq 10 ., As per Eq 9 , the individual forces in each muscle contribute to the overall stiffness of the limb while producing zero net torque at each joint to maintain equilibrium ., These isometric muscle forces have a metabolic cost , which we calculate as the sum of squares of muscle forces 39:, e n e r g y = ∑ k = 1 6 ( F m a x k * a k ) 2 ( 11 ) We simulate synergies that have been experimentally observed in a previous EMG study of static postures similar to those used during arm reaching tasks 19 ., These synergies couple the bi-articular muscles with the mono-articular elbow muscles as shown in Fig 4 ., Quantitatively , that study found that the elbow stiffness from co-contraction of the bi-articular muscles was approximately one half of the elbow stiffness from the mono-articular elbow muscles ., They did not find mono-articular shoulder muscles to have synergies with the bi-articular muscles ., Consequently , for our model , the activation of the shoulder synergy , ashoulder , activated the two mono-articular shoulder muscles with unity weight ., The activation of the the elbow synergy , aelbow , activated the mono-articular elbow muscles with unity weight and the biarticular muscles with weights of one-half ( Fig 4 ) ., In the presence of these synergies , one parameter suffices to change the orientation and shape of the endpoint stiffness ellipse: the ratio of elbow synergy activation to shoulder synergy activation , aelbow/ashoulder ., Increasing the activation of both synergies simultaneously and proportionately only increases the size of the ellipse but not its shape or orientation ( i . e . , the angle from the x-axis to the major axis of the ellipse ) ., As we will see in the results , this one-dimensional manifold in muscle activation space does not allow the realization of the arbitrary endpoint stiffness ellipses because the synergies , by coupling muscles , also couple two important stiffness characteristics: the stiffness ellipse’s shape and orientation ., That is , in the presence of synergies , as just described , there is only one free parameter that can be varied to control these characteristics , aelbow/ashoulder ., Therefore , changing orientation independently of shape is impossible ., To further explore this coupling of task constraints by synergies , we vary the ratio of shoulder synergy activation to elbow synergy activation ( by varying aelbow/ashoulder ) over a range of 2 orders of magnitude ( 1/10 to 10 ) to see how much the orientation of the ellipse is able to change ., In the absence of synergies in our model ( i . . e , all muscles can be activated independently ) , we can determine if the arm is able to meet the constraints that, all activation vectors lie within a unit 6-dimensional cube in the positive octant of the activation space 35 , 40 ( i . . e , all activations lie between 0 and 1 ), 0 ≤ a i ≤ 1 ( 12 ), the net joint torque vector is zero ( i . e . , the posture is in equilibrium ), R F m a x a → = 0 ( 13 ), and the endpoint stiffness has a given desired shape and orientation , as in Eq 10 J ˜ - T R ˜ F m a x a → = K ˜ e n d , d e s i r e d ( 14 ), Then if ∃a → s . t . Eqs 14 , 13 and 12 , are satisfied , K ˜ e n d , d e s i r e d is realizable in the absence of synergies ., An illustration of these constraints , the existence of a solution , and the potential for energy minimization in the absence of synergies is illustrated in Fig 5 for a simple 3-muscle model ., We use three muscles because this allows us to visualize the feasible activation space in 3D , and each of the linear constraints can be shown as a plane , whose intersection is a line that still holds some redundancy ., Since this example is for a manipulator with only one joint , the endpoint stiffness is only in the x-direction ., Thus, the feasible activation set begins as a 3-dimensional unit cube in the positive octant ., The constraint of zero endpoint force is a 2-dimensional plane in activation space passing through the origin ., This is because endpoint forces have a minimal value of 0 at zero activation ) ., The constraint for desired endpoint stiffness of unity is also a 2-dimensional plane in activation space , but it does not pass through the origin ., This is because muscle activation is required to produce stiffness: at the origin , there is no muscle contraction , therefore there is no muscle stiffness stiffness or endpoint stiffness ., This geometric interpretation 34 , 35 , 43 , 44 helps us understand the effect of synergies as additional constraints on feasible activations ., The intersection of the first two functional constraints is a one-dimensional linear subspace of solutions that mathematically satisfy Eqs 14 and 13 ., Further constraining this subspace by the activation N-cube ( Eq 12 ) results in the muscle activation solution space to realize a unity stiffness ., In this the feasible activation space , that has the structure of a one-dimensional subspace ( i . e . , a line ) , energy ( measured by the sum of the squares of the muscle forces , or concentric spheres ) can be minimized or maximized by varying the activation point in the feasible activation set ( i . e . , a point along the line ) ., The presence of even a simple synergy for this model ( a1 = 2a2 ) results in an additional constraint plane that passes though the origin that will reduce the feasible activation set , reducing the dimensionality of the solution space ., In this simple example , the dimensionality is reduced to zero—a unique solution 40 , 43 , 45 ., But even in high dimensions 41 , synergies will reduce what is already a well-structured space ., In our 6-muscle model , the activation hypercube is 6-dimensional ., The constraint of zero endpoint force is a 4-dimensional hyperplane in activation space passing through the origin ( 6 dimensions − 2 equilibrium constraints = 4-dimensional solution space; Eq 12 is a system of two equations , one for each joint ) ., The constraint for a desired endpoint stiffness is a 3-dimensional hyperplane in activation space ( 6 dimensions − 3 stiffness constraints = 3-dimensional solution space; as per Fig 3 , Eq 14 is a system of three equations , one for each unique element of the symmetric matrix K ˜ e n d , d e s i r e d ) ., The intersection of these two hyperplanes is a one-dimensional linear subspace ( 6 dimensions − 2 equilibrium constraints − 3 stiffness constraints = 1-dimensional feasible activation space ) embedded in 6-dimensional space ., It satisfies Eqs 14 and 13 ., If any part of this solution subspace lies in the activation N-cube ( satisfying Eq 12 as well ) , then the desired stiffness is realizable ., Furthermore , synergy constraints can reduce the dimensionality of the solution space to zero ( i . e . , there is a solution , it will be the unique solution of a point at the intersection of a line with a plane ) , or they can overconstrain the problem , making the desired stiffness unrealizable ., Within this context , we can now explore the range of achievable endpoint stiffness ellipse orientations given the arm posture and a desired ellipse shape ., To this end , we fixed both the condition number of the stiffness matrix and the posture , and then determined a set of desired endpoint stiffnesses , each corresponding to a different ellipse orientation ., We formulated a constrained quadratic programming problem , with the optimization criteria being minimizing the sum of squares of muscle activations ., If an optimum was found , then the orientation ( for that specific posture and ellipse shape ) is realizable ., We did this every 5° around the full range of orientations ( i . e . , 180° ) and then checked the fraction of these orientations that are realizable ., An example of the fraction of realizable orientations for all postures in the workspace is shown in Fig 6 ., The constraints in the realizability tests have five equality constraints ( Eqs 14 and 13 ) ., Since there are 6 muscles , if there is any solution which satisfies Eq 12 , in general there will be a one-dimensional feasible activation space for the desired endpoint stiffness embedded the 6-dimensional muscle activation space ., Vertex enumeration algorithms can be used to determine the vertices of this one-dimensional manifold ( which is a convex set 35 ) ., However , we and the available literature , are also interested in the maximal and minimal energy expenditures within this feasible activation space ., Therefore , we can use opposite quadratic programming optimization criteria to determine both of these energy expenditures ., For the minimal energy expenditure , as already described , our optimization criteria is to minimize the sum of squares of the muscle forces ., For maximal energy expenditure , our optimization criteria is to maximize the sum of squares of the muscle forces ., From these extreme values we can then determine the maximal amount of energy reduction that is possible ., For example , if the maximal energy expenditure is 0 . 5 , say , and the minimal energy expenditure is 0 . 35 , then there is a maximum of 30% reduction in energy possible ., Our rationale for quantifying these ratios is that , for given observed stiffness ellipsoid in human subjects experiments , we want to know whether or not the central nervous system could minimize energy expenditure ., If there is a large possible range of energies expended for a same endpoint stiffness ellipse , then it may only be possible for experimental means such as EMG to reach strong conclusions about energy minimization ., But if the range is low , then EMG measurements may not have the resolution to reveal much additional information about energy expenditure ( above the information obtained by only measuring the stiffness ellipse ) ., Fig 7 shows the fraction of realizable endpoint stiffness ellipse orientations for various ellipse shapes throughout the workspace ., We can make a couple of observations from Fig 7 ., First , posture has a very large effect on the range of realizable orientations ( also observed in 22 ) ., Second , the range of realizable orientations decreases with increasing ellipse eccentricity ., Thus a more uniform ellipse that is closer to a circle is easier to achieve throughout the workspace , but also arguably less able to set specific directions of higher or lower stiffness ., Also , our computational results for ellipse eccentricity = 1 is identical to the theoretical result determined by 8 ., To explore the effect of synergies in detail , we performed a more detailed analysis for a single posture ., In that sample posture , Fig 8 shows the range of sizes and orientations of the stiffness ellipse achievable when varying the ratio of elbow to shoulder synergy activations from 10−1 to 10 ., The arm endpoint is in a sample x − y position ( 0 , 1 ) , where each link of the arm has length of 1 ., The area of the ellipses in Fig 8 are normalized to be equal to each other to highlight the covarying shape and orientation of the stiffness ellipses ., The range of orientations is approximately 70° , which represents a realizable fraction of orientations of about 0 . 39 ., In this posture , shown in Fig 8 , for all 3 ellipse shapes , the fraction of realizable orientations is 1 ( all orientations are achievable ) in the absence of synergies ( Fig 7 ) ., In addition , we see that as the orientation of the ellipse in Fig 8 changes , the shape of the ellipse must also change ., The range of physically-realizable ratios of elbow to shoulder synergy activation are likely much less extreme than two orders of magnitude , which would result in an even smaller range of possible ellipse orientations ., Therefore , we see that using the synergies observed by Gomi and Osu 19 severely limits the ability of the CNS to control the shape and orientation of the endpoint stiffness of the arm ., Fig 9 shows the greatest possible reduction in energy expenditure given a stiffness ellipse shape and arm posture for any orientation ., Note the strong dependence on the posture of the arm ( i . e . , location in the workspace ) ., In general , the maximal possible energy reduction for many of the workspace postures for these stiffness ellipse shapes is low ( 10–30% ) , but can increase significantly to around 50% for some specific postures ., Fig 10 summarizes our findings , and compares them to prior work ., We see that implementing fewer synergies ( i . e . , fewer muscle groupings , that reflect greater correlation among muscle activations ) reduces the independent controllability of the size , shape and orientation of the stiffness ellipses , as well as the energy consumption ., The literature on muscle synergies is large and growing ., There are already several papers debating their origins , advantages , and disadvantages 24 , 28 , 29 , 42 , 46 ., The goal of this study , however , is to speak to the need pointed out by several authors to investigate the relationship between muscle synergies and the neuromechanical constraints that define a task ( sometimes also called task variables ) 24 , 28 , 34 , 40 , 42 , 46 ., We do so in the context of the neuromechanical consequences of using synergies while meeting the multiple and compounding constraints that define tasks in the ‘real world’ 34 , 41 , 45 , 47 , such as the well accepted need to regulate the stiffness of the endpoint of the arm ( e . g . , 1–22 ) ., In the literature mentioned above , the origins of synergies as well as their specific structure and permanence continue to be debated ., In their paper on static arm postures , Osu and Gomi ( 1999 ) mention that other arm synergies have been reported and that the regulation of muscle activation in static conditions seems to be quite different from that during movements ., Nevertheless , this does not affect our main finding that synergies—regardless of their origin , structure or permanence—have important neuromechanical consequences in a variety of functional domains ., This is because synergies imply a loss of control degrees of freedom ( i . e . , fewer independently controllable muscles ) ., Therefore , the specifics of the synergies we chose to simulate as reported by Osu and Gomi do not affect the generality of our results ., In fact , we went on to simulate five additional synergies as shown in Figs 11 and 12 , labeled Cases 2 through 6 ., In all cases , synergies yield a reduction in the controllability of the size , shape and orientation of the stiffness ellipses ., It is important to mention that comparing muscle coordination and stiffness regulation in static versus dynamic movement conditions may not be advisable—or even possible ., This stems from the fact that the physics and neuromuscular physiology of the control of static force versus movement are inherently distinct and can even be incompatible; see 36 , 45 , 48–50 and references therein ., In addition to the differences in their governing equations and the force-length properties of muscle , the control of movement at a neurological level additionally requires the careful and time-sensitive orchestration of alpha-gamma co-activation and reciprocal inhibition of eccentrically contracted muscles to prevent the disruption of the movement 51 ., This stems from the fact that the control of tendon excursions is overdetermined ( few joint angles determine the necessary lengths of all musculotendons ) ., This is the opposite of the underdetermined control of joint torques ( many combinations of muscle forces can equivalently produce a given net joint torque ) 36 , 41 ., Therefore , orchestrating alpha-gamma co-activation and reciprocal inhibition to produce movement imposes additional time-varying constraints that distort and reduce the feasible activation set for a given endpoint stiffness ellipse compared to the static condition ., From this perspective , our results for static endpoint stiffness are a best-case scenario as the additional constraints to produce movement will likely exacerbate the limitations imposed by synergies ., Understanding muscle coordination and stiffness regulation in static versus dynamic movement conditions remains an important area in motor control in need of attention 36 , 48 ., The feasible activation set—i . e . , all feasible neural commands to achieve a given task 34 , 35 , 41 , 43 , 44—has a well defined structure given by the biomechanics of the limb and the constraints defining the task ., Muscle synergies reduce the number of independent degrees of freedom for control from the ( usually large ) number of independently controlled muscles , to a smaller number of independently controlled groupings of muscle activations ., The presence of synergies , by reducing the number of independent degrees of freedom for control , naturally reduces the size and affects the shape of the feasible activation set—and therefore the set of tasks that are possible 34 , 40 ., This geometric approach uses a 6-muscle arm model with experimentally derived synergies to show that synergies severely constrain the ability to control the properties of the stiffness of the arm’s endpoint ., Furthermore , it also shows reduction in the flexibility of energy consumption to implement them ., That is , by reducing the dimensionality of the feasible activation set , synergies drastically limit the ability to orient the endpoint stiffness ellipse independently of its shape ., The range of achievable orientations in the absence of synergies is already very sensitive to posture , but still allows significant energy minimization in some postures ., Implementing synergies drastically reduces , and can even remove , the ability to minimize energy ., We would like to point out an important difference in our formulation compared with other modeling studies for arm stiffness 3 , 5 , 7 , 8 , 11 , 15 ., The general form of the joint stiffness matrix in these studies is ( for all equal moment arms ) :, K j o i n t = K s + K b K b K b K e + K b ( 15 ), where Ks is the shoulder stiffness provided by co-contraction of the mono-articular shoulder muscles , Kb is the bi-articular joint stiffness provided by co-contraction of bi-articular muscles , and Ke is the elbow stiffness provided by co-contraction of mono-articular elbow muscles ., This implies 3 constraints , and therefore 3 degrees of freedom for the system ., However , our formulation without synergies has 4 degrees of freedom since only 2 constraints must be satisfied ( R F 0 a → = 0 ) ., This study analyzes the extent to which the active stiffness ( i . e . , not including passive or reflexive muscle stiffness ) of the endpoint of a simulated arm can be controlled in the presence or absence of muscle synergies ., That is , the extent to which the endpoint of the limb would displace passively in response to a force perturbation in every direction ., That stiffness is the product of the level of activation of each muscle , and the anatomy and posture of the limb ., We assumed , as others have in the past , that the active stiffness of a muscle is linear and proportional to the maximal force a muscle can produce and the level to which it is activated ., While the linearity of active muscle stiffness with respect to muscle strength and activation is likely not entirely realistic , we focused on the effects of the presence or absence of muscle synergies ., Using a nonlinear relationship would likely produce different numerical results for the precise shape , size and orientation of stiffness ellipses ., However , it would not overcome the limitations that muscle synergies impose because those limitations come about from a reduction of the number of individually controllable muscles ., That is a matter of scale rather than quality ., Future work should naturally explore whether or not more realistic physiological mechanisms for muscle stiffness exacerbate the effects of synergies—particularly in neurological conditions ., In addition , this model is limited in that it does not take into account passive muscle stiffness , reflexive stiffness , or feedback pathways , which can clearly be used to minimize energy further depending on the frequency content of a perturbation or motor noise during a task ., It has been suggested 9 that some studies involving endpoint stiffness analysis may incorporate active reflex contributions 1 , 5 , 21 ., If only active , neurally-driven , stiffness properties are considered and there is no net force at the endpoint , then the endpoint stiffness matrix is symmetric ., It has been noted that any non-symmetric component of endpoint stiffness “can only be due to heteronymous inter-muscular feedback” 8 ., Although future work is needed to explore these effects , our study is still able to help shed light on conflicting findings even if we only considered active stiffness without producing any net endpoint force or torque ., A subtle but important issue is that studying symmetric endpoint stiffness does not take away from our findings , bur rather enhances our result about the functional limitations of synergies ., Adding a net endpoint force ( or torque ) deforms the symmetry of endpoint stiffness , but it also further constrains the range of stiffness modulation ., Balasubramanian and colleagues have made this point well by indicating that defining an endpoint force imposes an additional set of functional constraints that compromise the modulation of endpoint stiffness 34 ., Simila

Introduction, Methods, Results, Discussion

Much debate has arisen from research on muscle synergies with respect to both limb impedance control and energy consumption ., Studies of limb impedance control in the context of reaching movements and postural tasks have produced divergent findings , and this study explores whether the use of synergies by the central nervous system ( CNS ) can resolve these findings and also provide insights on mechanisms of energy consumption ., In this study , we phrase these debates at the conceptual level of interactions between neural degrees of freedom and tasks constraints ., This allows us to examine the ability of experimentally-observed synergies—correlated muscle activations—to control both energy consumption and the stiffness component of limb endpoint impedance ., In our nominal 6-muscle planar arm model , muscle synergies and the desired size , shape , and orientation of endpoint stiffness ellipses , are expressed as linear constraints that define the set of feasible muscle activation patterns ., Quadratic programming allows us to predict whether and how energy consumption can be minimized throughout the workspace of the limb given those linear constraints ., We show that the presence of synergies drastically decreases the ability of the CNS to vary the properties of the endpoint stiffness and can even preclude the ability to minimize energy ., Furthermore , the capacity to minimize energy consumption—when available—can be greatly affected by arm posture ., Our computational approach helps reconcile divergent findings and conclusions about task-specific regulation of endpoint stiffness and energy consumption in the context of synergies ., But more generally , these results provide further evidence that the benefits and disadvantages of muscle synergies go hand-in-hand with the structure of feasible muscle activation patterns afforded by the mechanics of the limb and task constraints ., These insights will help design experiments to elucidate the interplay between synergies and the mechanisms of learning , plasticity , versatility and pathology in neuromuscular systems .

The manner in which the nervous system coordinates the multiple muscles in the body is complex ., It has been studied for decades , but a more full understanding is needed to enable the development of effective evaluation and treatment methods in disorders that cause neuromuscular disability such as cerebral palsy and stroke ., In addition , the computational control of robots has and will continue to improve as the brain’s methods of muscular control are progressively reverse-engineered ., Here , we study the capacity of arm muscles to regulate the stiffness of the hand for tasks such as using tools , stabilizing hand-held objects , and using doors ., Using a simplified but generalizable model , we show that there will be necessary trade-offs in the functional capabilities of the limb if the nervous system chooses to control muscles in functional groups ., This adds to our understanding of the consequences of different strategies to control muscles for real-world tasks with multiple and often competing demands ., It enables future research and clinical experiments on the learning and execution of the multiple tasks of varying difficulty encountered in real life ., It also sheds light on the design of control strategies for robots to operate in human and unstructured environments .

stiffness, mechanical properties, medicine and health sciences, ellipses, nervous system, limbs (anatomy), geometry, biomechanics, mathematics, materials science, elbow, musculoskeletal mechanics, muscle physiology, musculoskeletal system, arms, anatomy, central nervous system, physiology, biology and life sciences, material properties, physical sciences, shoulders

journal.pgen.1004648

Genome-Wide Discovery of Drug-Dependent Human Liver Regulatory Elements

Adverse reactions to drug treatment constitute a substantial health problem that is a leading cause of morbidity and mortality in hospitalized patients 1 ., Differential expression of drug metabolizing enzymes and drug transporters is a major determinant of inter-individual drug response variability 2–5 ., By sequestering and metabolizing drug compounds in the liver and intestine , these enzymes and transporters effectively determine whether target organs and tissues are exposed to optimal drug dosages ., Several coding mutations in these proteins have been detected which lead to adverse outcomes 6–10 and reduced drug activity 11 , 12 ., Regulatory elements , including promoters and enhancers , also likely play an important role that has so far been largely uncharacterized 13 , 14 ., The systematic identification of drug-responsive regulatory elements would thus provide a unique resource to discover novel genetic variants that lead to differences in drug response ., The vast majority of pharmaceutical compounds are metabolized by the cytochrome P450 family ( CYP ) of enzymes ., Of these , CYP3A4 is the most abundantly expressed in sites of drug disposition in the liver 15 and is also thought to be responsible for the metabolism of at least 50% of prescribed pharmaceuticals 16 ., CYP3A4 activity can vary 5–20 fold between individuals 17 and its mRNA expression can vary as much as 120 fold 18 ., Only a few single nucleotide polymorphisms ( SNPs ) in the immediate CYP3A4 locus have been found to be associated with CYP3A4 hepatic expression 19–21 , suggesting that its variable expression could be caused by other genes and distant regulatory elements ., CYP3A4 is one of many targets of the nuclear receptor PXR ( coded by NR1I2 ) , which is expressed predominantly in the liver and intestine 22 and is essential for activating Phase I and II enzymes in response to xenobiotics ., PXRs broad substrate specificity allows it to be activated by a wide variety of drugs including the antibiotic rifampin , the malaria resistance drug artemisinin , the hypolipidemic agent mevastatin , and the chemotherapeutic agent paclitaxel 2 ., Relatively little is known about the mechanism by which PXR drives CYP3A4 transcription in vivo , although PXR response elements have been identified in the putative CYP3A4 promoter 23 and upstream cis-regulatory elements 24 , 25 that drive its expression in vitro ., Additional PXR responsive enhancers have been found for other CYPs 26 , 27 ., Chromatin immunoprecipitation followed by sequencing ( ChIP-seq ) of PXR-bound DNA elements in livers from mice treated with PCN ( a mouse PXR agonist ) identified >3 , 000 drug-induced binding sites 28 ., ChIP-seq for other drug-associated transcription factors such as LXR , RXR and PPARA has also been carried out in mouse liver 29 ., However , the inherent drug metabolism differences between mouse and humans , in particular for PXR and the mouse homolog of CYP3A4 22 , 30 , 31 , hinder the ability to directly translate these results to humans ., To identify PXR-associated regulatory elements in a genome-wide manner , we carried out RNA sequencing ( RNA-seq ) and ChIP-seq with antibodies for PXR and three different enhancer marks on primary human hepatocytes treated with rifampin or vehicle control ., These included the E1A binding protein p300 ( EP300/p300 ) which has been used to identify functional enhancers in vivo with high success rates 32 , and two histone marks , H3K4me1 and H3K27ac ., H3K4me1 marks both poised and active regulatory regions 33 while H3K27ac was shown to selectively mark active regions 34 , 35 ., We identified thousands of sequences that had rifampin induced ChIP-seq peaks ., A reporter validation screen of proximal promoters associated with these peaks yielded only a few functional rifampin-dependent sequences ., A similar assay for distal enhancers resulted in the identification of several novel drug-dependent enhancers ., Analyses of nucleotide variants in selected sequences found a common African haplotype in the GSTA locus to possibly affect rifampin sensitivity ., Our RNA-seq analyses found several differentially expressed genes , the majority of which are known to be involved in drug response ., The number of differentially expressed genes using a p-value cutoff , after adjustment for multiple testing less than or equal to 0 . 05 , was 157 ( Table S1 ) ., Amongst them , 11 were CYPs , with the top differentially expressed gene being CYP3A4 , similar to our qPCR results ., Of the eleven differentially expressed genes identified by qPCR , seven ( 64% ) were also found to be differentially expressed by RNA-seq ., It is worth noting that two ( CYP2C9 , CYP2C19 ) of the four genes that didnt replicate in the RNA-seq data , showed a non-statistically significant induction by rifampin in our RNA-seq ., We observed a massive recruitment of PXR binding across the genome following rifampin treatment ., PXR-bound DNA fragments clustered into 1 , 158 discrete peaks with DMSO treated cells versus 6 , 302 after treatment with rifampin ( Figure 1A , Table S2 ) , with only 239 overlapping in both datasets ( Figure 1B ) ., Rifampin treatment led to a small increase in the percentage of promoters ( 25 . 6% versus 24 . 1% ) bound by PXR and a larger increase for intronic ( 35 . 18% versus 29 . 9% ) and exonic ( 5 . 3% versus 1 . 8% ) regions ( Table S2 ) ., In contrast , there was a reduction in the percentage of rifampin-induced PXR binding sites in intergenic regions ( 33 . 9% versus 44 . 2% ) ., An analysis of the location of rifampin-induced PXR peaks found them to be enriched at transcription start sites ( TSSs ) , but not at a particular location upstream or downstream to the TSS ( Figure S1 ) ., The binding of p300 was also more extensive after rifampin treatment , with 13 , 811 peaks compared to 10 , 253 in DMSO-treated cells ( Figure 1A , Table S2 ) ., There was a larger overlap between rifampin and DMSO treated ChIP-seq peaks compared to PXR , with 4 , 374 ( 31 . 7% ) in common between the two sets ( Figure 1B , Table S2 ) ., We also observed a change in the functional distribution of binding sites , with rifampin increasing the percentage of intronic ( 42 . 6% versus 36 . 3% ) and intergenic ( 38 . 0% versus 31 . 7% ) p300 binding versus a small change in exons ( 3 . 8% versus 2 . 6% ) and a reduction in promoter binding ( 15 . 5% versus 29 . 4% ) ( Table S2 ) ., This result was consistent with our observation that only 1 , 076 rifampin-induced p300 peaks overlapped rifampin-induced PXR peaks ( Table S2 ) ., In contrast to PXR and p300 , the distribution of histone marks was relatively stable , with about 49 , 000 enriched islands of H3K4me1 activity and about 40 , 000 H3K27ac islands in both treatments ( Figure 1A ) ., We also observed a large overlap between rifampin and DMSO treated H3K4me1 ( 82 . 07% ) and H3K27ac ( 87 . 89% ) enriched islands ( Figure 1B , Table S2 ) ., Combined , these results suggest that histone marks are more stable in response to rifampin treatment compared to PXR and p300 ., We next looked at overlaps between the different ChIP-seq peaks ., Amongst the 6 , 302 PXR rifampin treated peaks 1 , 037 ( 16% ) overlapped p300 and around half overlapped histone marks ( 3 , 553 for H3K27ac and 2 , 942 for H3K4me1 ) ., This was similar for PXR peaks in the DMSO treated cells ( Table S2 ) ., For p300 we observed a greater overlap with histone marks , with ∼70% of the peaks overlapping either H3K27ac ( 9 , 487/13 , 811 ) and H3K4me1 ( 9 , 840/13 , 811 ) ., In the DMSO treated cells , we observed a much higher overlap for p300 peaks with the active H3K27ac mark ( 9 , 487/10 , 253; 92% ) versus H3K4me1 ( 7 , 789/10 , 253; 76% ) , suggesting that the p300 peaks in this condition tend to be in active regions ., There are multiple examples of promoter nucleotide variants that are associated with inter-individual drug response 4 , 5 , 13 , 38 , 39 ., We thus sought to identify drug-dependent promoters in our dataset which may harbor common variants with novel effects on drug response ., We selected 227 promoters for 200 genes ( some genes had more than one promoter; Table S3 ) from the LightSwitch Promoter Collection ( SwitchGear Genomics ) for genes whose expression was induced by rifampin or reside near rifampin-induced ChIP-seq peaks ( Table S3 ) ., Of the 227 promoters , 154 overlap a PXR peak , 45 overlap p300 , 164 overlap H3K27ac and 84 overlap H3K4me1 ( Table S3 ) ., This library consists of ∼1 , 000 bp proximal promoter fragments cloned into pLightSwich_Prom vector ( see Methods , Table S3 ) ., We also included two positive controls:, 1 ) The beta-actin promoter ( ACTB ) , a strong constitutive promoter that should not be induced by rifampin and, 2 ) The CYP3A4 proximal promoter , which is known to be induced by rifampin ., We tested the 227 promoters in HepG2 cell lines co-transfected with human PXR and treated with rifampin or vehicle control ( DMSO ) ., Out of the 227 tested promoters , 179 were found to be functional promoters ( >2 fold luciferase activity above empty vector ) in the DMSO treated cells ( Figure 2A , Table S3 ) ., Among those promoters , only 10 exhibited >2 fold increase in promoter activity upon rifampin treatment including our CYP3A4 proximal promoter control ( Figure 2A ) ., To confirm that the effects of rifampin were mediated through PXR , we also tested the 10 rifampin-induced promoters ( including CYP3A4 ) in a similar assay , only this time without co-transfecting human PXR , and found only 2 of them to be induced by rifampin ( >2 fold increase in promoter activity upon rifampin treatment ) and at much lower levels ( Figure 2B , Table S3 ) ., In addition , the CYP3A4 promoter was also not induced by rifampin in this assay ., In both experiments , our ACTB control was a strong promoter , but not induced by rifampin ., The overall lack of rifampin-sensitive promoters and previous results finding a role for enhancers in driving this drug response 24–27 suggests that other regulatory sequences , such as enhancers , may be involved in driving the effects of rifampin treatment on gene expression ., Since most of the promoters tested in our assay did not demonstrate increased activity in the presence of rifampin , we broadened our search for inducible regulatory elements to include enhancers ., To be more stringent in our analyses , we selected regions across the genome which showed PXR rifampin-induced binding in addition to all three enhancer marks ., For both the DMSO and rifampin treatments , we generated a merged track of all four marks , with each region in the track overlapping one to four peaks/island ., If , for example , a p300 peak is near a PXR peak , but they dont overlap , while both overlap a H3K4me1 and/or a H3K27ac island , they were considered all as one region ., Only 225 such regions were present in the DMSO treatment , while 1 , 387 were identified in the cells treated with rifampin ( Figure 3A ) ., Of the latter group , 1 , 297 regions were exclusive to the rifampin treatment and termed Rifampin-Induced Regions ( RIRs ) for downstream analyses ., CYP3A4 is by far the most well studied target of PXR , with well characterized regulatory sequences: the proximal promoter , a −7 . 5 kb upstream xenobiotic responsive enhancer module ( XREM ) 24 , and a −11 kb constitutive liver enhancer module of CYP3A4 ( CLEM4 ) 25 ., A second potential XREM , putatively regulating CYP3A7 , was additionally identified intergenically between CYP3A7 and CYP3A4 26 ., Our ChIP-seq data completely recapitulates this picture of regulation in primary human hepatocytes , with two large RIRs encompassing multiple rifampin-induced peaks ( Figure 3B ) ., It is also worth noting that the CYP3A4 locus is one of the few in which we observed a substantial difference in rifampin-induced enrichment of the H3K4me1 and H3K27ac marks ., To identify enriched biological pathways and functions within the set of 1 , 297 RIRs , we carried out a genomic analysis using the Genomic Regions Enrichment of Annotations Tool ( GREAT 40 ) ., Our top enriched term ( p-value 1 . 85×10−9; binominal fold enrichment ) , originating from Pathway Commons ( http://www . pathwaycommons . org ) , was ‘xenobiotics’ ( Figure 3C ) ., This was attributed to RIRs residing near the following genes: ABCB4 , ACSL1 , ADH1A , ADH6 , AKR1C2 , AKR1C3 , ALDH1A1 , CNDP2 , CYP26A1 , CYP2A6 , CYP2B6 , CYP2C19 , CYP2C8 , CYP2C9 , CYP2W1 , CYP3A4 , CYP3A7 , CYP4F12 , CYP4F3 , CYP7A1 , GCLC , GCLM , GSTA2 , GSTO1 , GSTO2 , HNF4A , MAT1A , MAT2A , MGST2 , MGST3 , NCEH1 , NNMT , PAPSS2 , PTGIS , SLC35D1 , SULT1B1 , SULT2A1 , UGDH , UGT1A1 ., In addition , we observed significant ( FDR adjusted p-value≤0 . 05 ) gene ontology enrichment terms for drug catabolic processes and other terms fitting with drug response ( Figure 3C , Table S4 ) ., We performed a similar analysis for RIR neighboring genes using QIAGENs Ingenuity Pathway Analysis ( IPA , QIAGEN Redwood City , www . qiagen . com/ingenuity ) and found enrichment for the PXR/RXR Activation Canonical pathway ( Figure S2 , Table S5 ) ., Combined , these results suggest that our RIRs are enriched near drug-associated genes ., Although ChIP-seq is a valuable tool for the identification of putative regulatory elements , functional studies are essential for the validation of such sequences ., We selected forty-nine putative enhancer sequences for validation using two different selection criteria:, 1 ) Forty-two RIRs residing near drug-associated genes as manually determined by the literature ., 2 ) Seven rifampin-induced PXR ChIP-seq peaks harboring SNPs that are in linkage disequilibrium ( LD ) with pharmacogenomics or drug related genome-wide association study ( GWAS ) lead SNPs , termed GWAS linked peaks ( GLPs ) ( Table S6 ) ., Previous studies have shown that several GWAS SNPs reside near potential regulatory elements that encompass SNPs that are in LD with the lead GWAS SNP 41–43 ., To increase our chances to identify these regulatory elements , we used our rifampin treated PXR-ChIP-seq dataset , instead of the RIRs , since it had a larger number of peaks ., SNPs in LD with pharmacogenomics GWAS hits were significantly enriched near rifampin-dependent PXR peaks compared to SNPs in LD with non-pharmacogenomic GWAS hits ( p<0 . 0001 , Chi-squared test ) ., None of the sequences chosen overlapped promoter regions ( i . e . were within −2500/+500 bp of a TSS ) ., Candidate enhancer sequences were cloned into the pGL4 . 23 ( Promega ) enhancer assay vector , which contains a minimal promoter followed by the luciferase reporter gene ., Since the peaks ( or islands ) enriched for the two histone marks are relatively long ( average length ∼5 kb , Table S2 ) , we selected shorter sequences within each RIR that encompass only the PXR/p300 peaks along with additional flanking sequence , up to 500 bp on each side of the peak ., We also used different positive controls:, 1 ) The ApoE liver enhancer 44 whose activity should not be enhanced by rifampin ., 2 ) A CYP3A4 promoter-enhancer combination ( p3A4-362 ( 7836/7208ins ) 24 ( Table S7 ) whose activity should be increased by rifampin treatment ., All constructs were tested for their enhancer activity in HepG2 cells transfected with human PXR and treated with either rifampin or DMSO , as previously done for the promoter screen ., Out of the 49 sequences tested , 19 ( 38 . 7% ) showed significant reporter expression levels versus the empty vector ( ≥2 two fold ) in either condition: 15 RIRs and 4 GLPs ( Figure 4A , Table S7 ) ., Among these 19 positive enhancers , we observed three types of enhancers:, 1 ) Seven enhancers that were active at similar levels with DMSO and rifampin , termed ‘rifampin independent’ ., 2 ) Five enhancers that were active without rifampin , but whose expression levels significantly increased upon rifampin treatment , termed ‘rifampin increased’ ., 3 ) Seven enhancers that were active only when treated with rifampin and were called ‘rifampin dependent’ ., Two of these ( GLP1 and GLP2 ) are located in the CYP2C locus ( Figure S3 ) and contain SNPs in LD with pharmacogenomics GWAS SNPs for warfarin maintenance dose 45 , 46 , acenocoumarol maintenance dose 47 and response to clopidogrel therapy 48 ., Combined , these results show that enhancer activity can be modulated by rifampin ., We next determined whether common nucleotide variation within functional , drug-dependent enhancers could alter their activity ., For these experiments we selected five enhancers that were either rifampin increased or dependent and near important drug-response genes ., These included RIR7 , which overlaps the putative CYP3A7 XREM ( Figure 3B; chr7: 99339411–99341549; hg19 ) 26 and was rifampin dependent in our assays ., We also selected RIR46 , which is located in the glutathione S-transferase alpha ( GSTA ) locus near GSTA2 ( chr6: 52609942–52611507; hg19 ) ( Figure S4 ) and was rifampin-increased in our assays ., The GSTA family of enzymes are known to be involved in the metabolism of various xenobiotics 49 ., We also selected three different GLP sequences: GLP1 , 2 , and 5 ., Both GLP1 ( chr10:96507473–96508107; hg19 ) and GLP2 ( chr10:96696182–96696970; hg19 ) are located in the CYP2C locus ( Figure S3 ) , which harbors several CYP metabolizing enzymes and has been analyzed extensively in various pharmacogenomic studies ., GLP5 ( chr2:234672744–234673398; hg19 ) harbors a single SNP , rs3771341 , that is in LD with several GWAS lead SNPs correlated with altered bilirubin levels 50–53 and was rifampin-increased in our study ., This element is located in the UDP glucuronosyltransferase 1 family , polypeptide A cluster ( UGT1A ) , ∼4 kb upstream of the UGT1A1 transcription start site ( Figure S5 ) ., UGT1A enzymes have important roles in the metabolism of xenobiotics and both coding and promoter variants within them have been associated with adverse drug reactions 54 ., We determined common haplotypes in all five sequences using the phased 1000 Genomes data ( Table S8 ) ., Common haplotypes for all five sequences ( Table S8 ) were then cloned into our enhancer assay vector ( pGL4 . 23 ) either by amplifying DNA from individuals from various ethnic backgrounds from the studies of pharmacogenomics in ethnically diverse populations ( SOPHIE ) cohort 55 or by site-directed mutagenesis and sequence verified ., The sequences were then tested for enhancer activity in HepG2 cells transfected with human PXR and treated with either rifampin or DMSO , and compared to the ancestral haplotype ., Out of the five tested enhancers , one haplotype in RIR46 showed a substantial difference in enhancer activity ., For RIR46 , we observed 1 . 85-fold ( 4 . 01/2 . 16 ) increase in response to rifampin for haplotype 3 , however after adjusting for multiple testing the variance in the response was too high to be significant ( adjusted p-value\u200a=\u200a0 . 095 by FDR; ANOVA; Figure 4B , Table S8 ) ., This haplotype is present at a frequency of 6 . 7% in the 1000 Genomes AFR population ., It is worth noting that our success rate in finding haplotypes that significantly alter enhancer activity was low ., Nonetheless , the observation that a haplotype in RIR46 could possibly affect enhancer activity suggests that nucleotide variants in these enhancers could lead to differential enhancer activity ., By carrying out RNA-seq and ChIP-seq on rifampin and DMSO treated human hepatocytes , we have uncovered numerous drug-dependent regulatory elements ., We observed that promoters bearing a rifampin-dependent signature were largely unable to independently induce the expression of a reporter upon rifampin treatment ., This result suggested that other gene regulatory elements , such as enhancers , could constitute the predominant group of target sequences that are activated by this drug ., An analysis of nucleotide variation in these enhancers showed that specific variants could affect enhancer activity , raising the possibility that nucleotide variants in enhancers could contribute to pharmacogenomic phenotypes more broadly ., Our RNA-seq analyses that combined eight different individuals identified 157 differentially expressed genes , using a p-value cutoff that adjusted for multiple testing less than or equal to 0 . 05 ( Table S1 ) ., Many of these genes are known to be involved in drug response and the top differentially expressed gene was CYP3A4 , fitting with its role as the most abundantly expressed gene in sites of drug disposition in the liver 15 ., The number of differentially expressed genes ( 157 ) , using our 0 . 05 cutoff , was much lower than the number of RIRs ( 1 , 297 ) and PXR and p300 rifampin treated ChIP-seq peaks ., This difference could be attributed to having multiple regulatory elements regulating the same gene ., It could also be due to our conservative RNA-seq differential expression cutoff , which if relaxed would increase our gene list ., Another cause for this can be that several of our identified peaks are not functional regulatory elements ., Our functional characterization for example , found only 19 of the 49 tested sequences ( 38 . 7% ) to have significant reporter expression levels versus the empty vector ( ≥2 two fold ) ., Of note though , that this could also be due to the different cell types and conditions that were used in this assay , as later described ., Similar studies have been conducted to detect environmentally induced regulatory elements 28 , 56–59 ., For example , testing for changes in estrogen receptor α and RNA polymerase II occupancy due to estradiol , tamoxifen or fulvestrant treatment in MCF-7 cells , found differences in ligand regulation with tamoxifen leading to downregulation while fulvestrant increased RNAPII occupancy 58 ., In our study , we observed a large rifampin-induced recruitment of PXR and p300 binding across the genome ., This was particularly notable in the CYP3A locus , which was radically altered by rifampin treatment ., For the most part , the regulatory hotspots that we identified in this region are consistent with established literature ., It is worth noting , however , that we saw almost equal recruitment of PXR and p300 to the XREMs that putatively regulate CYP3A4 and CYP3A7 , despite the fact that we observed substantially less CYP3A7 mRNA induction by qPCR and RNA-seq ., We did not observe large changes in the assayed histone marks , H3K4me1 and H3K27ac ., This could be attributed to these sites being poised to be activated by various xenobiotic responses , though H3K27ac has been previously shown to mark active enhancers 34 , 35 ., To get at these differences more systematically , we performed a second analysis in which H3K4me1 and H3K27ac peaks were called for the rifampin treatment using the DMSO treatment as the control ( Table S9 ) ., We observed 110 differential islands for H3K27ac , and only 2 for H3K4me1 ( one of which was in the CYP3A4 locus ) ., The fact that we observed more rifampin-induced peaks for H3K27ac is consistent with its hypothesized role in marking active enhancers ., Combined , these results suggest that while these two marks are incredibly stable in the face of massive changes in PXR/p300 binding , there are still measurable drug-induced changes in chromatin structure ., Previous reports have found enhancers to be responders of drug treatment 24–27 ., Our functional validation results broadly suggest that promoters on their own are largely incapable of driving the effects of rifampin induction ., Instead , our results imply that enhancers appear to be induced by rifampin , and through their interaction with promoters , drug response genes are activated ., We tested over 200 promoters of genes based on relatively loose selection criteria: that their respective genes either showed increased expression following rifampin treatment , were tagged by our rifampin ChIP-seq peaks , or were near them ., On the other hand , for enhancers we required that candidates had all four marks ., While it is possible that our promoter selection criteria missed out on important drug response promoters , we would still expect to achieve a higher rifampin induction success rate than what we observed in our assays ( 10/227; 4 . 4% ) ., While our assays may have not been ideal for these purposes , i . e . testing promoters in vitro in PXR-transfected HepG2 cells , both our negative and positive controls showed the expected results ( Figure 2A ) ., Furthermore , rifampin-induced promoters tested in non-PXR transfected conditions showed much lower induction by the drug ( Figure 2B ) ., Further systematic assays including ones carried out in vivo will be needed to better address this hypothesis ., While 12 out of the 19 functional sequences identified by our screen exhibited basal enhancer activity , 7 sequences exhibited enhancer activity only upon rifampin treatment ( Figure 4A ) ., These sequences would not have been identified by conventional ChIP studies conducted in physiological conditions , nor would they be validated in functional assays without drug treatment ., Together our results suggest that ChIP-seq datasets are dependent on the environmental conditions in which they were performed , and that there are likely many hidden enhancers which only become active following a specific stimulus ., We identified several functional enhancers that were rifampin-increased or rifampin-induced whose location was near pharmacogenomic-associated variants ., One of these elements is RIR46 , which resides near GSTA2 , a Phase II enzyme involved in the detoxification of numerous drugs 49 , and is thus a likely target ., Coding variants in GSTA2 have been shown to affect its detoxification efficiency 60 , 61 and promoter variants in GSTA2 have been suggested to affect its expression levels 62 , 63 ., We identified a haplotype present in the 1000 Genomes AFR population that confers increased rifampin sensitivity ( Figure 4B ) ., Both GSTA2 and GSTA1 ( which is 28 kb downstream to GSTA2 ) , have been shown to have important roles in catalyzing carcinogenic substrates and nucleotide variation in them has been shown to be associated with cancer 60 , 64 ., Previous work has shown a significant difference in the distribution of coding SNPs in both these genes between African Americans and Caucasians 60 ., However , these coding SNPs are not associated with prostate cancer disease status 60 , suggesting that other factors could be playing a role ., Future studies could examine whether variants in RIR46 or other RIRs are associated with these phenotypes ., It is worth noting that there are several caveats to our study ., Our ChIP-seq experiments only analyzed hepatocytes from a single donor at a single time point ( 24 hours post treatment ) , selected for being commonplace in rifampin induction studies ., It is possible that there are enhancers that play a role much earlier than 24 hours post-treatment ., To test regulatory sequences in reporter assays , it was also necessary to clone smaller fragments from each RIR bearing the PXR and p300 peaks ., It is possible that we missed flanking sequences that were essential for enhancer function ., Furthermore , because we had a limited amount of primary hepatocytes from the same donor , our reporter assays were carried out in PXR-transfected HepG2 cells , which could result in false negatives ., Finally , we employed the pGL4 . 23 vector , which is a commonly used enhancer assay vector , but possesses a very short TATA-box containing minimal promoter that may not be compatible with all of the enhancer sequences tested ., Promoter regions upstream of drug metabolizing enzymes and transporters are presumed to be the major targets of xenobiotic activators of PXR , such as rifampin ., Our work challenges this conception and strongly supports the idea that the direct targets could be enhancer elements , which may subsequently interact with promoters to enhance gene transcription ., Combined , our results show that ChIP-seq in combination with drug treatment has a large potential in identifying novel regions in the genome associated with drug response ., These regions can provide exceptional candidates for the detection of nucleotide variants associated with inter-individual differences in drug response ., Our methodology could easily be adapted to other drugs/target/tissue combinations ., With whole genome datasets becoming more widely used as a clinical toolbox , the ability to highlight these important drug response regions in the genome is of extreme importance ., Cryopreserved human hepatocytes from a 19 year old Caucasian male donor with no history of medications ( Lot Hu8080 , Life Technologies ) were thawed in CHRM recovery media ( Cat# CM7000 , Life Technologies ) and cultured in CHPM plating media ( Cat#CM9000 , Life Technologies ) on 6-well collagen-coated plates ( Life Technologies ) ., After 6 hours , the media was swapped with maintenance media , consisting of phenol-free Williams E media containing culture incubation supplements ( Cat#CM4000 , Life Technologies ) and 0 . 01% DMSO or 10 µM rifampin ( Sigma ) for 24 hours ., The rifampin dose ( 10 µM for 24 hours ) was used based on previously reported assays achieving high induction rates in human hepatocytes 65 , 66 ., Dexamethasone , which is included separately with the culture incubation supplements , was not added to the media as it can activate PXR ., Cultured hepatocytes were treated with DMSO or rifampin for 24 hours in triplicate ., The cells were then washed with PBS , and lysed directly with Buffer RLT from the RNAeasy mini kit ( Qiagen ) with the on-column DNase digestion step ., One µg of total RNA was used to generate cDNA using the RT2 First Strand Kit ( Qiagen ) ., Gene expression levels for 84 genes of interest was determined using the “Drug Metabolism: Phase I Enzymes” RT2 Profiler PCR Array ( Qiagen ) ., Five housekeeping genes ( B2M , HPRT1 , RPL13A , GAPDH , ACTB ) were used to control for loading ., Fold induction was calculated using the ΔΔCt method 67 ., Total RNA was acquired as described for qPCR for two replicates each of DMSO and rifampin treated hepatocytes ., Libraries were made with ScriptSeq v2 RNA-Seq Library Preparation Kit ( Epicenter ) ., Briefly , 3–5 ug of total RNA were subjected to rRNA removal using Ribo-Zero Magnetic Kit ( Epicenter ) prior to library construction ., 5 ng of the rRNA-depleted sample was fragmented enzymatically and annealed with random hexamer to create the first strand of cDNA ., Upon removal of the RNA template transcript by RNase , Terminal-Tagging Oligo ( TTO ) , a known 5′-sequence tag , a 3′-random sequence , and a terminally blocked 3′ end to prevent priming of DNA synthesis , was added to create cDNA with known sequence tags at their 5′ and 3′ ends for directionality ., Upon purification , adaptors with barcodes were added to cDNA fragments and enriched by 15 cycles of PCR ., Sequencing was carried out on an Illumina HiSeq ., The resulting reads were demultiplexed and aligned to the human genome ( hg19 ) using TopHat v2 . 0 . 10 68 ., Read counts for each gene in the RefSeq annotation were obtained using NGSUtils 69 so as to allow comparison to the RNA-seq data from the other 7 primary hepatocytes treated with and without rifampin and sequenced with SOLiD as described in 37 ., Analysis for differential expression across the nine replicates was performed using DESeq2 70 ., DESeq2 was chosen due to its capability in handling multifactorial experimental designs , in this case treated with rifampin versus control and SOLiD sequences versus Illumina ., DESeq2 was then used to perform a likelihood ratio test between the model of treatment plus sequencing type versus the simplified model of just sequencing type in order to identify genes differentially expressed upon treatment with rifampin ., Twelve million cells ( an entire 6 well plate ) per immunoprecipitation were fixed with 1% formaldehyde for 15 min and quenched with 0 . 125 M glycine .,

Introduction, Results, Discussion, Materials and Methods

Inter-individual variation in gene regulatory elements is hypothesized to play a causative role in adverse drug reactions and reduced drug activity ., However , relatively little is known about the location and function of drug-dependent elements ., To uncover drug-associated elements in a genome-wide manner , we performed RNA-seq and ChIP-seq using antibodies against the pregnane X receptor ( PXR ) and three active regulatory marks ( p300 , H3K4me1 , H3K27ac ) on primary human hepatocytes treated with rifampin or vehicle control ., Rifampin and PXR were chosen since they are part of the CYP3A4 pathway , which is known to account for the metabolism of more than 50% of all prescribed drugs ., We selected 227 proximal promoters for genes with rifampin-dependent expression or nearby PXR/p300 occupancy sites and assayed their ability to induce luciferase in rifampin-treated HepG2 cells , finding only 10 ( 4 . 4% ) that exhibited drug-dependent activity ., As this result suggested a role for distal enhancer modules , we searched more broadly to identify 1 , 297 genomic regions bearing a conditional PXR occupancy as well as all three active regulatory marks ., These regions are enriched near genes that function in the metabolism of xenobiotics , specifically members of the cytochrome P450 family ., We performed enhancer assays in rifampin-treated HepG2 cells for 42 of these sequences as well as 7 sequences that overlap linkage-disequilibrium blocks defined by lead SNPs from pharmacogenomic GWAS studies , revealing 15/42 and 4/7 to be functional enhancers , respectively ., A common African haplotype in one of these enhancers in the GSTA locus was found to exhibit potential rifampin hypersensitivity ., Combined , our results further suggest that enhancers are the predominant targets of rifampin-induced PXR activation , provide a genome-wide catalog of PXR targets and serve as a model for the identification of drug-responsive regulatory elements .

Drug response varies between individuals and can be caused by genetic factors ., Nucleotide variation in gene regulatory elements can have a significant effect on drug response , but due to the difficulty in identifying these elements , they remain understudied ., Here , we used various genomic assays to analyze human liver cells treated with or without the antibiotic rifampin and identified drug-induced regulatory elements genome-wide ., The testing of numerous active promoters in human liver cells showed only a few to be induced by rifampin treatment ., A similar analysis of enhancers found several of them to be induced by the drug ., Nucleotide variants in one of these enhancers were found to alter its activity ., Combined , this work identifies numerous novel gene regulatory elements that can be activated due to drug response and thus provides candidate sequences in the human genome where nucleotide variation can lead to differences in drug response ., It also provides a universally applicable method to detect these elements for other drugs .

biotechnology, medicine and health sciences, functional genomics, health care, genome analysis, genetic elements, dna, gene regions, molecular genetics, pharmacoeconomics, health economics, gene expression, biochemistry, gene regulatory networks, genetics, biology and life sciences, genomics, computational biology, pharmacogenomics, human genetics

journal.pgen.1001294

Epistatic Interaction Maps Relative to Multiple Metabolic Phenotypes

An epistatic interaction between two genes occurs when the phenotypic impact of at least one of the genes is dependent on the other 1 ., This dependence is often a consequence of an underlying functional relationship between the two genes 2 , 3 ., By extending the study of epistasis from individual interactions to networks of interactions , recent work in S . cerevisiae has demonstrated that genome-wide patterns of epistasis can be used to uncover the global organization of biological systems 4–7 ., In such studies epistatic interactions are identified as instances where the effect of a double perturbation on growth differs from the expectation based on the observed effects of the corresponding single perturbations 8 ., The choice of growth rate as a phenotype is motivated by the role of epistasis in the dynamics of selection 9 , and by the fact that growth rate , a proxy for fitness , can be accurately measured in a high-throughput manner 10 ., In parallel to the experimental efforts , large-scale studies of epistasis on growth phenotypes have also been pursued computationally , especially using the approach of flux balance analysis 6 , 11 , 12 ., Such computational studies have offered preliminary novel insight before the availability of corresponding experimental data , e . g . in predicting a coherence principle ( monochromaticity ) in the organization of epistatic interaction networks 6 , subsequently observed experimentally 7 , 13 ., Overall , large-scale studies of epistasis have become increasingly relevant to functional genomics 4 , 7 , 14 , drug development 13 , 15 , 16 and evolutionary biology 17 , 18 ., Albeit important , growth rate is just one of many possible phenotypes relative to which genes can interact epistatically with each other ., In contrast with the rapidly increasing understanding of the nature and scope of epistatic interactions relative to growth , many questions remain unresolved with respect to epistasis relative to non-growth phenotypes ., Are interactions relative to non-growth phenotypes as widespread as interactions with respect to growth ?, Do genes tend to interact relative to more than one phenotype , and if so , is the type of epistasis consistent across phenotypes ?, How much more dense can an epistatic network become upon adding new phenotypes ?, Do interactions with respect to specific phenotypes provide biological insight than cannot be obtained from knowing interactions relative to growth rate ?, Most importantly , does the potential presence of multi-phenotype epistasis affect the way cells operate and evolve ?, While these questions have not , to our knowledge been asked before , epistasis relative to non-growth phenotypes is not in itself a new concept ., Interactions between polymorphisms have been detected by using multiple mRNA transcript levels as phenotypes 19 ., Another recent study searched for interactions among genes conferring resistance to a DNA-damaging agent and showed that a denser network was observed with respect to the capacity to cope with the damaging agent , than was found with respect to growth rate under standard conditions 20 ., In addition , in the study of human genetic diseases , while epistasis relative to disease-related traits poses challenging technical problems , it is a potentially important component , especially in light of the relative paucity of explanatory power detected through the analysis of individual loci 21–23 ., Hence interactions relative to diverse phenotypes are likely widespread and informative ., However , the combinatorial complexity resulting from the large number of possible genetic perturbations and phenotypes has prevented so far a systematic analysis of the extent and biological implications of this phenomenon ., In this work we report the computational study of epistatic interactions in a flux balance model of metabolism that is simple enough to allow an exhaustive computation of all possible perturbations relative to all possible phenotypes , but at the same time realistic enough to provide meaningful biological insight ., Specifically , we use an experimentally informed variant of the method of minimization of metabolic adjustment ( MOMA ) in a genome-scale metabolic network model of Saccharomyces cerevisiae 24 to predict all steady state metabolic reaction rates ( fluxes ) in response to all possible single and double enzyme gene deletions ., By comparing single and double mutant values for all fluxes and defining appropriate metrics , we construct an epistatic map for each flux phenotype ( Figure 1 ) ., This multi-phenotype genetic interaction map allows us to explore for the first time the properties and significance of epistasis across a combinatorial set of perturbations and phenotypes ., Quantifying epistasis relative to multiple metabolic flux phenotypes introduces three fundamental challenges , one specific to the use of flux balance models and two broadly relevant to any study of multi-phenotype epistasis ., The first issue is the reliability of flux predictions for deletion mutants ., The availability of experimentally determined growth phenotypes for all gene deletion mutants in S . cerevisiae has allowed for extensive evaluation of the yeast models capacity to predict mutant growth ., These previous studies 25 , 26 , including a comparison of model predictions against experimental growth measurements for 465 gene deletion mutants under 16 metabolically diverse conditions 26 , have demonstrated that the yeast model can predict deletion mutant viability with high accuracy ., Furthermore , observed discordances between model predictions and experimentally determined mutant growth phenotypes have been used in refinements of the existing yeast model , further bolstering the ability of the model to accurately mimic the effect of different gene deletions 25 , 27 ., In addition to effectively predicting single mutant growth , flux balance models have also been shown to predict viabilities of double deletion mutants with high accuracy 28 ., However , while model predictions of mutant growth have been evaluated extensively , comparisons between measured and predicted fluxes through the underlying metabolic reactions in different mutants are less readily available 29 ., To address this need we recently evaluated the ability of the yeast model to predict the fluxes through central carbon metabolism in single gene deletion mutants by comparing model predictions to a previously released compendium of experimentally measured mutant fluxes 30 ., An assessment of different approaches for mutant flux prediction revealed that an experimentally driven variant of the minimization of metabolic adjustment 31 gives the best correlation with measured fluxes ( Spearman rank correlation greater than 0 . 90 , Figure S1 ) , and hence chose it for our calculations ( See Materials and Methods ) ., In essence , this method implements the hypothesis that the metabolic response to genetic perturbation will be a minimal rerouting of flux around the insult ., A conceptual illustration of the methodology for predicting mutant fluxes is shown in Figure 2 , with a detailed quantitative description provided in the Materials and Methods and Text S1 ., A second issue , which is critical to any multi-phenotype study of epistasis , is the choice of a metric for quantifying epistasis ., The quantification of epistasis requires an assumption as to how the phenotypic effects of non-interacting mutations combine: deviations from this expectation are inferred to be indicative of epistasis ., While previous work has provided both theoretical and empirical evidence for how the effects of mutations on fitness combine 8 , no comprehensive study has yet explored how the effects of mutations on metabolic fluxes combine ., To this end we evaluated two standard metrics for computing epistasis ( multiplicative and additive definitions ) , in addition to a novel metric ., This novel metric was designed so as to avoid making any assumption on how the phenotypic effects of two mutations combine ., Avoiding such assumptions is ideal for detecting epistasis across multiple phenotypes , relative to which the effects of mutations may combine differently ( See Text S1 and Table S1 ) ., However , comparing these three different quantitative definitions ( See Materials and Methods and Text S1 ) , we found that epistasis relative to metabolic fluxes is overall robustly detectable independent of the metric used ( Figure S2 ) ., In the following analyses , based on this result , a multiplicative model is used , and all main conclusions were verified to be robust relative to different metrics ., A third issue arising in a global analysis of epistatic effects with respect to metabolic fluxes is the partitioning of interactions into different classes of epistasis 32 ., These different classes of epistasis represent different ways in which the combined effect of two mutations may defy expectation , and can be indicative of different types of underlying functional relationships between genes 2 , 6 ., In moving from growth to flux phenotypes , the classification of interactions becomes more complex , due to the fact that fluxes can increase or decrease upon genetic perturbation , while the growth rate typically only decreases ., While the increased complexity present in our data allows for discrimination of many different classes of interactions ( See Figure S3 and Text S1 ) , for the current analysis we consolidate all sub-classes of interactions into two groups , synergistic and antagonistic ( See Materials and Methods , Figure S3 ) ., A synergistic interaction between two genes indicates that the change in the observed flux ( phenotype ) caused by the simultaneous deletion of both genes is greater than expected based on the effects of the corresponding single deletions , while an antagonistic interaction indicates a flux change in the double mutant that is less than expected ., Synergistic interactions are indicative of a compensatory relationship between two genes , such that the extreme phenotype of the double mutant is a consequence of this compensation being lost ., Antagonistic interactions are indicative of two genes working together towards some function , such that the reduced phenotypic effect of the double mutant occurs because the common function is compromised by the loss of either of the genes individually ., These preparatory steps allowed us to compute and analyze a 3-dimensional epistatic map for the yeast metabolic network , as illustrated in Figure 1 ., The complete set of synergistic and antagonistic epistatic interactions were reduced to a high-confidence set by independently applying a standard deviation cutoff to the distributions of epistasis relative to each phenotype ( See Materials and Methods and Figure S4 ) ., Considering only these high confidence interactions , it was found that 100 of the 672 genes in the model interact with respect to at least one of the 293 fluxes active under the modeled minimal glucose condition ., To simplify the subsequent analysis of the epistatic map , we consolidated the 100 interacting genes into the 30 metabolic processes to which they are assigned in the model , and counted an interaction between two processes if any gene from one process interacts with any gene from the other ., This consolidated epistatic map is represented in Figure 3A , where the total numbers of synergistic ( red ) , antagonistic ( blue ) and mixed ( yellow ) interactions between pairs of biological processes , across all phenotypes , are displayed as a stacked histogram ., Mixed interactions between two processes occur when some pairs of genes across the processes interact synergistically , while others antagonistically ., Figure 3A indicates that such mixed process interactions are less frequent than process interactions that are purely synergistic or antagonistic , suggesting that the previously observed monochromaticity of epistatic interactions between biological processes 6 applies to diverse metabolic phenotypes ., Monochromaticity is a consequence of the fact that genes in the same biological processes function cohesively , and hence share similar patterns of epistatic interactions 4 , 5 , 7 ., Notably , however , in our multi-phenotype epistatic map , the “color” ( synergistic or antagonistic ) of the interaction between two processes depends on the phenotype observed ., Figure 3B demonstrates that this dependence of process interaction colors on phenotype is due to the fact that individual gene pairs often interact synergistically relative to some phenotypes and antagonistically relative to others ., This pattern reveals that the class of an epistatic interaction is not an absolute characteristic of a pair of genes , but rather a characteristic of the gene-gene-phenotype triad ., This suggests that the functional relationship between two genes is not necessarily one dimensional , but may depend on the function ( the phenotype ) being probed ., The intuition that different phenotypes convey complementary insight into the functional associations between genes and processes was confirmed in a quantitative manner by determining how many unique interactions each phenotype contributes to the 3D epistatic map ., Figure 4 shows that the total number of interactions identified when considering all phenotypes is ∼8 times larger than can be identified relative to any individual phenotype , although the exact increase in interaction coverage is dependent on the threshold for defining a significant interaction ( See Materials and Methods ) ., Figure 4 also shows that 83 of the 293 total metabolic flux phenotypes are required to identify all unique epistatic interactions in yeast metabolism ., Examining the distribution of metabolic processes where these 83 phenotypes come from ( Figure S6 ) , reveals that they are spread across all metabolic processes ., This suggests that a set of phenotypes that represents all metabolic functions is required to identify all epistatic interactions ., Conversely , this implies that different phenotypes are providing insights into unique aspects of the functional relationships between genes ., To solidify the observation that different flux phenotypes reveal unique aspects of the functional relationships between genes , we next focus on the epistatic networks relative to two secretion phenotypes ( succinate , Figure 5B , and glycerol , Figure 5C ) ., We chose to focus on secretion flux phenotypes because they are the most tractable fluxes to measure experimentally , and hence potentially the most relevant for future experimental studies ., Both of these secretion flux epistatic networks contain several interactions that are not detected relative to the growth phenotype ( Figure 5A ) ., In particular , in the succinate secretion network , the genes that are part of complex II of the electron transport chain ( ETC II ) display synergistic interactions with several other biological processes ( Figure 5B ) ., Among these interactions , which are indicative of an unexpectedly large increase in succinate secretion in the double mutant , the one between serine biosynthesis and ETCII has been reported in previous experimental efforts to overproduce succinate 33 ., This interaction occurs because the predicted alternate pathway for serine biosynthesis produces succinate as a byproduct , and ETC II is the primary route through which this succinate is metabolized in the wild-type ( Figure 6A , Figure S7 ) ., Thus , interactions with respect to succinate may in general probe the way in which TCA cycle intermediates are produced and consumed ., In the glycerol secretion phenotype network there is enrichment for synergistic interactions between glutamate biosynthesis and respiratory processes ( Figure 5C ) ., Among these interactions , the interaction between glutamate synthase and the electron transport chain is supported by experimental data gathered in the context of ethanol production optimization 34 ., This epistatic interaction is a consequence of the fact that glutamate biosynthesis , the electron transport chain and glycerol biosynthesis correspond to three of the major routes for cytosolic NADH oxidation ( Figure 6B , Figure S9 ) ., Thus , interactions with respect to glycerol secretion may reflect the way in which different processes contribute to cellular redox balance ., These examples , and others in Text S1 and Figure S8 , further demonstrate that interactions with respect to metabolic flux phenotypes can provide detailed insights into different aspects of the functional relationships between genes ., So far , we have shown that epistatic interactions between gene deletions relative to metabolic flux phenotypes are ubiquitous , and can provide an understanding of the relationships between different processes in the cell ., The ubiquity of epistasis relative to metabolic flux phenotypes brought to our attention the possibility that these complex network-level functional interdependencies might impose constraints on evolutionary trajectories ., We hypothesized that this phenomenon might manifest itself in the form of increased evolutionary constraints on enzymes that are involved in many epistatic links with other genes ., Such a relationship between epistatic connectivity and evolutionary rate has been recently observed in the experimentally constructed global genetic interaction network with respect to growth rate in yeast 7 ., Thus , we set out to explore whether predicted connectivity with respect to metabolic phenotypes other than growth rate are also correlated with evolutionary constraint ., To this end we calculated the Spearman rank correlation between the number of interactions in which different genes participate and the evolutionary rates of such genes , as measured by their non-synonymous to synonymous substitution ratios ., This correlation was calculated separately for synergistic and antagonistic interactions relative to each of the 293 flux phenotypes , for a total of 586 correlations ., The distributions of correlation coefficients for synergistic and antagonistic interactions across all phenotypes are shown in Figure 7A ., Both distributions significantly deviate from zero , with an overall bias towards negative correlations ( Sign test , p\u200a=\u200a8 . 5×10−25 ( synergistic ) , 2 . 2×10−54 ( antagonistic ) , n\u200a=\u200a293 ) ., This trend towards negative correlations suggests that genes involved in many interactions with respect to metabolic flux phenotypes do indeed evolve slower ., While the negative skew of these distributions is robustly maintained upon removal of most potential confounding factors ( see Figure S11 ) we found that it is significantly reduced when controlling for the codon bias of the genes ( Figure 7B ) ., Codon bias is a proxy for gene expression level , which previous research has shown to be the dominant correlate of evolutionary rate 35 , 36 ., Therefore , we cannot rule out that a portion of the apparent evolutionary importance of genes with a high degree of genetic interactions across different phenotypes may be explainable by the expression level of the genes ., Yet , regardless of whether the interaction degree correlates with evolutionary rate or gene expression level , either result indicates the functional importance of these multi-phenotype hubs ., The increased expression level of these hubs in fact supports their central role in metabolic function ., Furthermore , in our model , epistatic interaction degree with respect to growth flux alone is not significantly anti-correlated with evolutionary rate , even without controlling for expression level ( Figure S12 ) ., This indicates that the importance of genes is associated with their total phenotypic impact , not just their impact on growth ., While the distributions of correlations between evolutionary rate , and both synergistic and antagonistic interaction count , shift towards zero when controlling for codon bias , the distribution remains significantly different from zero only for antagonistic interactions ( p\u200a=\u200a0 . 07 ( synergistic ) , p\u200a=\u200a2 . 3×10−31 ( antagonistic ) ) ., We believe that this observation can be understood by considering more closely the relationships between genes that interact antagonistically , versus synergistically ., An antagonistic interaction implies that the phenotypic effect of deleting a gene is reduced in the absence of its interaction partner ., A possible interpretation of this is that a genes full function , as manifested in its associated phenotypic effect , is contingent on the presence of its antagonistic interaction partner ., Therefore highly antagonistic genes are phenotypic hubs , whose evolutionary changes are constrained by the dependency of other genes upon them ., Conversely , the reduced constraint on synergistic hubs can be understood by considering that a synergistic interaction between two genes implies that the phenotypic impact of deleting a gene is increased in the absence of its interaction partner ., This can be interpreted as a genes function being compensated for by its synergistic interaction partners ., Therefore , the reduced correlation with evolutionary rate for synergistic hubs may reflect the fact that the phenotypic effect of changes in such hubs is dampened by the presence of their interaction partners ., The implications of the current analysis are not limited to yeast ., In fact , multi-phenotype epistatic interactions may be relevant to the manifestation and treatment of human disease ., Given the previously discussed importance of multi-phenotype hub genes , it is likely that perturbations of these genes would have major effects in a biological network ., Translating this observation to humans , we hypothesize that the disruption of more highly connected genes in the human metabolic network would be more likely to result in a disease state ., We sought evidence for this by evaluating whether the epistatic connectivity of genes in the yeast model was predictive of the role of their human homologs in genetic diseases ., Indeed , we observe a significant difference between the connectivity of yeast homologs of human genes that have been associated with a genetic disease , versus those that have not ( Figure 8 ) ., While the statistical significance is limited due to the small sample size , this result provides support for the growing sentiment that majority of human genetic disorders are a consequence of complex interactions between numerous cellular components 1 , 21 ., We described the systematic generation of epistatic interaction networks relative to all observable phenotypes in a genome-scale model of yeast metabolism ., Analysis of these networks revealed that different metabolic flux phenotypes yield different sets of interactions , and that a large set of phenotypes is required to capture all interactions ., The basis for these observations is that different metabolic flux phenotypes capture different aspects of gene function ., This is likely a consequence of the complex wiring of metabolic networks , which include multiple branching pathways , shared pools of commonly used metabolites and a high level of interconnectedness between different metabolic processes: seemingly remote processes on the metabolic chart may nonlinearly affect a third readout process ( the phenotype ) ., Furthermore , because of this complexity , the relationships between different genes and processes may not be easily captured by straightforward patterns , as indicated by the observation that the same genes can interact synergistically relative to some phenotypes and antagonistically relative to others ., From a functional genomic perspective , the results imply that , in future studies of epistasis , the set of observed phenotypes could be selected so as to influence the set of interactions identified and to maximize insight into the functional organization of the biological process of interest ., While the focus here has been on metabolism , this concept can be generalized to other types of biological networks ., For instance , mRNA transcript levels may be the most appropriate phenotype to tease out the logic of transcriptional regulatory networks 37 and phosphorylation states the most relevant for signal transduction pathways ., Furthermore , our results demonstrate that the particular mRNA levels or protein phosphorylation states monitored should depend on the particular regulatory module or transduction pathway of interest ., An additional layer of complexity that has not yet been addressed here is the dependence of epistatic networks on environmental conditions ., As hinted to before ( Supplementary Figure 3 in 6 ) epistatic networks will likely vary under different conditions ., Hence , future extensions of the current work may explore the complexity and significance of environmental conditions as a fourth dimension in the epistatic matrix of Figure 1 ., From an evolutionary perspective , we found that the number of epistatic interactions with respect to multiple metabolic flux phenotypes is strongly anti-correlated with the genes evolutionary rates and expression levels ., This anti-correlation is larger than found with the number of epistatic interactions relative to growth phenotype only ., On the surface this result seems surprising , given that growth rate and fitness are often taken to be synonymous with one another , and genes that have a large impact on fitness would be expected to evolve slower ., However , one must consider that growth in the model is based solely on the capacity to produce biomass components , while fitness in an organisms natural environment is assuredly more complex ., An organisms success ( in other words , its fitness ) likely depends on the complex interplay of a multitude of biological properties , including the proportions and efficiency of resources utilized , the choice of secreted byproducts ( which can influence the environment and the interactions with other species ) , and how fluxes are managed in the face of varying nutrient availability ., Thus , the apparently reduced importance of the growth flux , and conversely , the increased importance of all metabolic phenotypes , may simply be reflective of the relative simplicity of growth in the model , when compared to the complexity of growth in the wild ., More broadly , our results raise the possibility that the apparent robustness observed in the insulated environment of the laboratory may not translate to an organisms natural environment , where additional constraints exist with respect to not just how fast one grows , but the precise manner in which this is accomplished ., A potential limitation of past and present , computational and experimental studies of the evolutionary impact of epistasis may lie in the use of gene deletions as the mutation relative to which epistasis is detected ., While gene deletion mutations have been effective in terms of uncovering functional dependencies and the evolutionary constraints imposed by these dependencies , left unanswered is the evolutionary impact of epistasis relative to smaller perturbations to gene function ., It is these minor perturbations , such as those caused by amino acid substitutions or stochastic fluctuations in protein levels that the cell must constantly confront ., If epistasis relative to these small perturbations is as ubiquitous as has been observed relative to gene deletions , this begs the question as to how the cell copes with the complexity of a large number of long-distance nonlinearities affecting virtually every metabolic function ., While experimental studies have begun to address this question 38 , perhaps computational frameworks such as flux balance analysis can be used to extend these analyses to the genome-scale ., For flux balance analysis , or any computational framework , to adequately address this problem much work will have to be done to more fully understand the phenotypic consequences of small genetic perturbations ., While our current analysis is purely computational , we anticipate that xperimental measurements of interactions based on multiple metabolic phenotypes will be increasingly feasible and valuable in the near future ., Our analysis provides predictions about the properties of multi-phenotype epistatic networks , in addition to a plethora of specific interaction predictions to which these future experiments can be compared ( data downloadable at http://prelude . bu . edu/multi-phenotype-epistasis ) ., Finally , being the first genome-scale analysis of multi-phenotype epistatic networks , we hope that the groundwork we have laid with respect to quantifying , discretizing and analyzing multi-phenotype epistatic interaction networks will aid future experimental and computational studies using similar approaches to help unravel the functional complexity of biological systems ., To enable our study of multi-phenotype epistasis at a genome-scale we utilized flux balance models 24 ., Specifically , to compute steady state reaction rates ( the fluxes , vi ) in deletion mutants , we used the iLL672 yeast stoichiometric reconstruction ., Flux balance models take as input the stoichiometry of all known metabolic reactions in the modeled organism , along with possible constraints on flux ranges , and through a Linear Programming optimization step provide predictions of fluxes through each metabolic reaction ., The complete stoichiometry of an organism is typically represented mathematically as the stoichiometric matrix , S . Each row i of the matrix S represents a metabolite , and each column j represents a reaction , with an entry Sij representing the stoichiometric coefficient of metabolite i in reaction, j . The set of possible flux solutions is constrained by imposing a steady state assumption along with bounds on individual fluxes ., The set of steady state solutions is described mathematically as the null space of the matrix S , and dictates that the production of each metabolite is equaled by its consumption ., Bounds on individual fluxes are described by inequality constraints , and are used to model known limitations , such as nutrient availabilities , reaction reversibility and maintenance requirements ., Constraints on nutrient limitation in the present study were set so as to mimic as closely as possible the media conditions from a recent study by Blank et al . 30 ., This condition was selected since it allowed us to use experimental flux data from that study to perform more accurate flux predictions throughout our work ( See below ) ., Upon setting the linear constraints , a particular flux solution is typically computed by searching the optimal value of a given linear combination of the fluxes ., Previously utilized optimization criteria are maximal ATP production 39 , minimization of total flux 40 , and the most commonly used maximization of biomass production 41 ., Formally , this can be expressed as a Linear Programming ( LP ) problem:where cr is the coefficient of flux r in the objective function ( r\u200a=\u200a1 , … , R , with R is the total number of reactions ) , and αj and βj are the upper and lower bounds on reaction j , respectively ., Because of the nature of our study , accurate predictions for all individual fluxes are desirable ., Hence , we wanted our flux predictions to match available experimental data as closely as possible ., To this end we evaluated several optimization criteria for predicting fluxes in deletion mutants by comparing flux predictions to a previously released set of experimentally measured fluxes in yeast single deletion mutants 30 ( see Text S1 ) ., Our evaluation demonstrated that the most accurate optimization criteria utilized the previously described Minimization of Metabolic Adjustment ( MOMA ) criteria 29 , along with an experimentally constrained wild type solution 24 ., In effect , this criterion assumes that upon a gene deletion , fluxes will undergo a minimal rearrangement , compatible with the flux constraints imposed by the gene deletion ., We found that the per

Introduction, Results, Discussion, Materials and Methods

An epistatic interaction between two genes occurs when the phenotypic impact of one gene depends on another gene , often exposing a functional association between them ., Due to experimental scalability and to evolutionary significance , abundant work has been focused on studying how epistasis affects cellular growth rate , most notably in yeast ., However , epistasis likely influences many different phenotypes , affecting our capacity to understand cellular functions , biochemical networks adaptation , and genetic diseases ., Despite its broad significance , the extent and nature of epistasis relative to different phenotypes remain fundamentally unexplored ., Here we use genome-scale metabolic network modeling to investigate the extent and properties of epistatic interactions relative to multiple phenotypes ., Specifically , using an experimentally refined stoichiometric model for Saccharomyces cerevisiae , we computed a three-dimensional matrix of epistatic interactions between any two enzyme gene deletions , with respect to all metabolic flux phenotypes ., We found that the total number of epistatic interactions between enzymes increases rapidly as phenotypes are added , plateauing at approximately 80 phenotypes , to an overall connectivity that is roughly 8-fold larger than the one observed relative to growth alone ., Looking at interactions across all phenotypes , we found that gene pairs interact incoherently relative to different phenotypes , i . e . antagonistically relative to some phenotypes and synergistically relative to others ., Specific deletion-deletion-phenotype triplets can be explained metabolically , suggesting a highly informative role of multi-phenotype epistasis in mapping cellular functions ., Finally , we found that genes involved in many interactions across multiple phenotypes are more highly expressed , evolve slower , and tend to be associated with diseases , indicating that the importance of genes is hidden in their total phenotypic impact ., Our predictions indicate a pervasiveness of nonlinear effects in how genetic perturbations affect multiple metabolic phenotypes ., The approaches and results reported could influence future efforts in understanding metabolic diseases and the role of biochemical regulation in the cell .

An epistatic interaction between two genes occurs when the phenotypic impact of one gene is dependent on the other ., While different phenotypes have been used to uncover epistasis in different contexts , little is known about how cell-scale genetic interaction networks vary across multiple phenotypes ., Here we use a genome-scale mathematical model of yeast metabolism to compute a three-dimensional matrix of interactions between any two gene deletions with respect to all metabolic flux phenotypes ., We find that this multi-phenotype epistasis map contains many more interactions than found relative to any single phenotype ., The unique contribution of examining multiple phenotypes is further demonstrated by the fact that individual interactions may be synergistic relative to some phenotypes and antagonistic relative to others ., This observation indicates that different phenotypes are indeed capturing different aspects of the functional relationships between genes ., Furthermore , the observation that genes involved in many epistatic interactions across all metabolic flux phenotypes are found to be highly expressed and under strong selective pressure seems to indicate that these interactions are important to the cell and are not just the unavoidable consequence of the connectivity of biological networks ., Multi-phenotype epistasis maps may help elucidate the functional organization of biological systems and the role of epistasis in the manifestation of complex genetic diseases .

genetics and genomics/microbial evolution and genomics, evolutionary biology/microbial evolution and genomics, genetics and genomics/functional genomics, computational biology/metabolic networks, microbiology/microbial physiology and metabolism, computational biology/systems biology

journal.pgen.1005581

A Hereditary Enteropathy Caused by Mutations in the SLCO2A1 Gene, Encoding a Prostaglandin Transporter

The use of capsule endoscopy and balloon endoscopy has provided a better understanding of the features of small bowel ulcers among various gastrointestinal disorders , such as Crohn’s disease ( CD ) , intestinal tuberculosis , and nonsteroidal anti-inflammatory drug ( NSAID ) –induced enteropathy 1 , 2 ., Previously , we proposed a rare clinicopathologic entity characterized by multiple small intestinal ulcers of nonspecific histology and chronic persistent gastrointestinal bleeding as chronic nonspecific multiple ulcers of the small intestine ( CNSU ) 3 , 4 ., The macroscopic findings of CNSU are characterized by multiple thin ulcers in a linear or circumferential configuration and concentric stenosis , and apparently mimic those of NSAID-induced enteropathy 4–7 ., CNSU predominantly occurs in females and the symptoms , such as general fatigue , edema , and abdominal pain , appear during adolescence ., The clinical course of the disease is chronic and intractable with reduced effects of immunosuppressive treatment including prednisolone and azathioprine ., Although CNSU predominantly occurs in females , it also appears to be an autosomal recessive inherited disorder because of frequent parental consanguinity 8 ., To identify the causative gene for this disorder , we performed whole-exome sequencing and identified recessive mutations in the SLCO2A1 gene , encoding a prostaglandin transporter , as causative variants ., Furthermore , we replicated our findings in other patients with CNSU and established a genetic cause for this inherited disease ., We performed whole-exome sequencing in five affected females with CNSU ( A-V–2 , B-IV–3 , C-IV–3 , D-II–4 , and D-II–5 ) and one unaffected individual ( A-V–3 ) ( Figs 1 and 2 ) ., Parental consanguinity was noted in families A , B , and C . The average depth of sequence coverage in the whole-exome sequencing data was 68 . 9× ( S1 Table ) ., We identified a total of 368 , 403 variants , of which 20 , 271 were non-synonymous or splice-site mutations ., By filtering the data with dbSNP135 , we found 2 , 406 variants located in 1 , 578 genes ., Based on the parental consanguinity of the patients , we focused on the shared genes with homozygous variants among three affected individuals ( A-V–2 , B-IV–3 , and C-IV–3 ) and found nine candidate genes , PCSK9 , ASPM , DAG1 , SLCO2A1 , MCPH1 , EFEMP2 , DDHD1 , PKD1L3 , and SYNGR1 ., After consideration of the results for the unaffected individual ( A-V–3 ) and another two sibling patients ( D-II–4 and D-II–5 ) , only SLCO2A1 remained as a candidate gene ., The three patients with parental consanguinity ( A-V–2 , B-IV–3 , and C-IV–3 ) had a homozygous splice-site mutation in the SLCO2A1 gene , c . 1461+1G>C or c . 940+1G>A ( Fig 1 and Table 1 ) ., The two sibling patients had compound heterozygous mutations , c . 664G>A ( p . Gly222Arg ) and c . 1807C>T ( p . Arg603X ) ., All four mutations were predicted to be loss-of-function or damaging mutations by SIFT and PolyPhen–2 ., The four identified SLCO2A1 mutations were confirmed to be present in five affected individuals ( A-V–2 , B-IV–3 , C-IV–3 , D-II–4 , and D-II–5 ) by Sanger sequencing ( Fig 1 ) ., Segregation analysis of patient A-V–2 revealed that her unaffected parents , sister , brother , daughter , and son carried the heterozygous c . 1461+1G>C mutation ( Fig 1A ) ., To compensate for bias in our analysis , such as the possibility of ethnic-specific variants , we genotyped the four candidate variants in 747 unaffected Japanese subjects from our previous genome-wide association study 9 ., All mutations except for the c . 940+1G>A mutation were absent in controls ( Table 2 ) ., The c . 940+1G>A mutation was found in the heterozygous state in 3 of 747 controls , showing a similar allele frequency of 0 . 0022 to the public exome database for the Japanese population ( HGVD database ) ., Subsequently , we screened all 14 coding exons and intron-exon boundaries using Sanger sequencing in 12 other CNSU patients , and identified two novel mutations , c . 421G>T ( p . Glu141X ) and c . 1372G>T ( p . Val458Phe ) ( Table 1 ) ., Eleven of the 12 patients were found to have homozygous ( nine patients ) or compound heterozygous ( two patients ) SLCO2A1 mutations that were rare and predicted to be deleterious by SIFT , PolyPhen–2 , and PROVEAN ( Table 1 ) ., The remaining patient ( 66-year-old female ) , who did not have an SLCO2A1 mutation , was diagnosed as CNSU because of multiple ulcerations in the duodenum and jejunum ., Although anti-tumor necrosis factor-α antibody therapy was ineffective , clinical improvement was achieved by enteral nutrition ., Because CNSU can be misdiagnosed as CD in some cases , we searched for the six identified mutations of SLCO2A1 in CD patients to identify concealed CNSU patients ., Among 603 patients previously diagnosed as CD 10 , two individuals ( patients 17 and 18 in Table 1 ) were found to carry a pair of compound heterozygous SLCO2A1 mutations , c . 940+1G>A/c . 547G>A and c . 940+1G>A/c . 421G>T , respectively ., The c . 547G>A mutation ( p . Gly183Arg ) , was a novel mutation at a highly conserved site and predicted to be deleterious by in silico analysis ., The clinical information for the two individuals was reviewed retrospectively , and the diagnosis of CNSU was confirmed ., In total , we identified seven deleterious SLCO2A1 mutations in 18 patients ( Table 2 ) ., In total , we found 18 patients with CNSU confirmed by genetic analysis ( Table 1 ) ; 14 of them were female ., In all patients , the ulcers occurred in the ileum ( Fig 2A–2D ) ., The stomach and duodenum were affected in five ( 27 . 8% ) and eight ( 44 . 4% ) patients , respectively ., Because mutations in the SLCO2A1 gene , encoding a prostaglandin transporter , have been reported to be the pathogenic cause of primary hypertrophic osteoarthropathy ( PHO; OMIM 614441 ) 11–13 , we investigated whether CNSU patients had clinical manifestations of PHO ., Although no patients had any clinical manifestations of PHO requiring treatment , mild digital clubbing or periostosis was present in seven of 18 patients ( S2 Table ) ., Moreover , three male patients ( patients 12 , 16 , and 17 ) fulfilled the major clinical criteria for PHO , having digital clubbing , periostosis , and pachydermia ., There were no female patients who fulfilled the major clinical criteria ( Fig 2E and 2F ) ., Among the identified SLCO2A1 mutations , a splice-site mutation of intron 7 ( c . 940+1G>A ) was the most frequent , and nine of the 18 patients were homozygous for this mutation ., There were no obvious correlations between the types of mutations and the clinical phenotypes ., Because the SLCO2A1 gene encodes a prostaglandin transporter that mediates the uptake and clearance of prostaglandins , the urinary levels of prostaglandin E metabolite ( PGE-M ) were measured ., The urinary PGE-M levels in CNSU patients were significantly higher than those in unaffected individuals ( p = 0 . 00013; S1 Fig ) ., Using RT-PCR , we demonstrated that splicing of the SLCO2A1 mRNA , derived from biopsy specimens of the small intestine , was altered in affected siblings with the homozygous c . 940+1G>A mutation ( patients 6 and 7 ) compared with a control individual ( Fig 3A ) ., Sequencing of the RT-PCR products revealed deletion of the whole exon 7 of SLCO2A1 , leading to a frameshift at amino acid position 288 and introduction of a premature stop codon after six amino acid residues ( p . R288Gfs*7 ) ., Sequencing of the RT-PCR products of the transcripts in peripheral blood mononuclear cells from patient A-V–2 revealed that the homozygous c . 1461+1G>C mutation led to a 23-bp frameshift insertion into intron 10 , resulting in a premature stop codon ( p . I488Lfs*11 ) ( S2 Fig ) ., For functional analysis of the intact and truncated SLCO2A1 proteins , we investigated the 3H-labeled prostaglandin E2 ( PGE2 ) transport ability in HEK293 cells transfected with intact SLCO2A1 and mutant SLCO2A1 proteins for each identified mutation ( c . 940+1G>A , p . Gly222Arg , p . Arg603X , p . Glu141X , p . Val458Phe , and p . Gly183Arg ) ., HEK293 cells transfected with intact SLCO2A1 showed the ability for PGE2 transport ., In contrast , HEK293 cells transfected with the mutant SLCO2A1 proteins were unable to uptake 3H-labeled PGE ( p < 0 . 0001; Fig 3B ) ., These findings demonstrated that the mutant SLCO2A1 proteins identified in patients lost their functional ability as a PGE transporter ., In control sections of normal small intestinal mucosa , SLCO2A1 was expressed on the cellular membrane of vascular endothelial cells in the small intestine , as evaluated by immunohistochemistry and immunofluorescence staining with a specific anti-SLCO2A1 antibody recognizing the fifth extracellular domain coded by exons 9–11 of the SLCO2A1 gene ( Fig 3C ) ., We then analyzed the expression of SLCO2A1 in the small intestine of affected individuals with the homozygous c . 940+1G>A mutation ( patients 6 and 7 ) by immunofluorescence staining ., However , the immunofluorescence staining did not detect any SLCO2A1 protein in the vascular endothelial cells of the patients ( Fig 3C ) ., These results indicated that the entire SLCO2A1 protein was unexpressed in affected individuals with the homozygous c . 940+1G>A mutation , consistent with the results of the mRNA transcript sequencing ., To investigate the subcellular localization of SLCO2A1 and the truncated SLCO2A1 protein ( ΔSLCO2A1 ) corresponding to the c . 940+1G>A mutation , we constructed expression vectors for GFP-SLCO2A1 and GFP-ΔSLCO2A1 fusion proteins and transfected them into HEK293 cells ., GFP-SLCO2A1 was localized on the cellular membrane ( Fig 3D , arrows ) as well as in the cytoplasm of transfected HEK293 cells , while GFP-ΔSLCO2A1 did not accumulate on the cellular membrane ( Fig 3D ) ., In this study , we performed whole-exome sequencing in five Japanese patients with CNSU and one unaffected individual , and identified the SLCO2A1 gene as the candidate for this disorder ., We further confirmed that SLCO2A1 gene mutations were involved in the pathogenesis of CNSU by genotyping of control subjects and other CNSU patients ., Moreover , a genetic analysis of 603 patients previously diagnosed as CD revealed that two CNSU patients had been included in this disease group ., In total , we identified seven different mutations in the SLCO2A1 gene , comprising two splicing-site mutations , two truncating mutations , and three missense mutations , as the causative gene defects for CNSU ., Therefore , we propose a more appropriate nomenclature , “chronic enteropathy associated with SLCO2A1 gene” ( CEAS ) , for this disease ., The SLCO2A1 gene encodes a prostaglandin transporter that may be involved in mediating the uptake and clearance of prostaglandins in numerous tissues 14–16 ., This gene has already been reported as a causative gene for a subtype of PHO 11 ., In fact , three of the seven identified mutations , c . 664G>A , c . 940+1G>A , and c . 1807C>T , have also been reported as causative mutations for PHO 11–13 , 17 , 18 ., We found that three male patients with CEAS had all of the major clinical features of PHO , such as digital clubbing , periostosis , and pachydermia ., Moreover , either digital clubbing or periostosis was present in seven of 18 patients ., These findings indicate that CEAS and PHO share a causative gene and that their clinical features are profoundly influenced by other modifiers ., Taken together with the facts that that CEAS predominantly occurs in females and PHO predominantly occurs in males 8 , 17 , a sex-influenced gene or hormone may be the main disease modifier ., Zhang et al . 17 reported that two female family members of a PHO patient had no clinical features of PHO , despite having a homozygous SLCO2A1 mutation ., Moreover , it is interesting to note that these two siblings both had anemia and hypoalbuminemia , suggesting that they had CEAS ., PHO is also known to be caused by mutations of HPGD , encoding 15-hydroxyprostaglandin dehydrogenase ( 15-PGDH ) , as well as SLCO2A1 19 ., The transmembrane prostaglandin transporter encoded by the SLCO2A1 gene delivers prostaglandins to cytoplasmic 15-PGDH , resulting in their degradation 14 , 20 ., Because 15-PGDH is the main enzyme for prostaglandin degradation , systemic PGE2 levels are increased in patients with HPGD deficiency ., Consistent with the findings in our present investigation , Zhang et al . 11 reported that the urinary levels of PGE2 and PGE-M in SLCO2A1-deficient individuals with PHO are significantly higher than those in controls ., In fact , the clinical features of PHO were assumed to be caused by excessive PGE2 ., Meanwhile , although elevated levels of PGE2 in gastrointestinal tissues are commonly known to protect against mucosal inflammation via the prostaglandin receptor EP3/EP4 21–23 , multiple intestinal ulcers occur in CEAS ., This discrepancy and the pathogenesis of intestinal ulcers need to be clarified in future studies ., Although CEAS is presumed to be unaccompanied by immunological inflammation in its pathogenesis , a portion of CEAS patients can be misdiagnosed as CD because of the shared common clinical features , such as multiple small intestinal ulcers , anemia , and hypoalbuminemia ., In this study , two of 603 patients previously diagnosed as CD were found to be affected with CEAS by genetic analysis ., Because corticosteroid and anti-tumor necrosis factor-α antibody therapies are ineffective for CEAS , recognition and precise diagnosis of CEAS are critical to avoid unnecessary therapies ., The findings of our investigation lead us to conclude that genetic analysis in addition to detailed clinical information including digital clubbing , blood tests , and gastrointestinal examinations are invaluable for distinguishing CEAS from CD ., Cases of a similar enteropathy referred to as cryptogenic multifocal ulcerous stenosing enteritis ( CMUSE ) have been reported in Western populations 24–26 ., This enteropathy has been shown to be an autosomal recessive inherited disease caused by mutations in the PLA2G4A gene 27 ., CEAS and CMUSE share common clinicopathologic features with respect to age of onset , chronic and recurrent clinical course , and nonspecific stenosing small intestinal ulcers 4 , 25 ., However , the sex predominance , response to steroid therapy , and lesion sites are obviously different between the two conditions ., The PLA2G4A gene encodes cytoplasmic phospholipase A2-α ( cPLA2α ) , which catalyzes the release of arachidonic acid from membrane phospholipids ., CMUSE patients with compound heterozygous mutations of PLA2G4A have been reported to show globally decreased production of eicosanoids such as PGE2 and thromboxane A2 , resulting in multiple ulcers of the small intestine and platelet dysfunction 27 , 28 ., Because impaired prostaglandin use underlies CEAS , CMUSE , and NSAID-induced enteropathy , we propose a new entity of gastrointestinal disorders , namely “prostaglandin-associated enteropathy” ., In conclusion , we have identified loss-of-function mutations in the SLCO2A1 gene as the cause of CEAS ., The present findings clearly indicate that CEAS is a genetically distinct entity independent of other gastrointestinal disorders including CD , NSAID-induced enteropathy , and CMUSE ., Further studies are needed to elucidate the pathogenesis of CEAS and identify new therapeutic molecular targets for “prostaglandin-associated enteropathy” ., Written informed consent for genetic studies was obtained from each individual ., The study was approved by the institutional review board at each collecting site in accordance with the Declaration of Helsinki Principles ., We obtained blood samples and family pedigrees from 17 Japanese patients with CNSU and eight unaffected family members in 15 families ., The diagnosis of CNSU was based on the published clinical criteria and clinical courses ( S3 Table ) 8 , 29 ., Genomic DNA samples from 747 participants in our previous genome-wide association study for ulcerative colitis 9 and 603 patients with CD 10 , 30 were used after excluding subjects who recalled their consent ., DNA was extracted from peripheral blood using standard methods ., Whole-exome sequencing in five affected individuals ( A-V–2 , B-IV–3 , C-IV–3 , D-II–4 , and D-II–5 ) and one unaffected individual ( A-V–3 ) was performed to identify candidate genetic variants ( Fig 1 ) ., Genomic DNA was enriched using a TruSeq Exome Enrichment Kit ( Illumina , San Diego , CA , USA ) according to the manufacturer’s instructions , and paired-end sequencing was carried out with an Illumina HiSeq 2000 instrument ., Reads were aligned to the human genome reference sequence ( hg19 NCBI build 37 . 1 ) and decoy sequences using BWA software 31 ., Duplicate reads were removed with the Picard program ( http://picard . sourceforge . net/ ) ., Recalibration and realignment of the data were accomplished with Genome Analysis Toolkit ( GATK ) 32 , 33 ., Single nucleotide variants and small insertions and deletions ( indels ) were identified by GATK Unified Genotyper ., The effect of each missense mutation was predicted using SIFT ( http://sift . jcvi . org/ ) 34 , PolyPhen–2 ( http://genetics . bwh . harvard . edu/pph2/ ) 35 , and PROVEAN ( http://provean . jcvi . org/ ) 36 ., To compensate for bias in our analysis , such as the possibility of ethnic-specific variants , we genotyped the four candidate variants identified by exome sequencing in 747 unaffected Japanese subjects by Sanger sequencing and restriction fragment length polymorphism analysis ( S4 Table ) ., For further validation , Sanger sequencing of all exons of the SLCO2A1 gene in other CNSU patients was performed using standard protocols ., Finally , we genotyped the six identified mutation sites in the SLCO2A1 gene in clinically diagnosed CD patients , because CNSU can be misdiagnosed as CD ., Urine samples were collected from 15 CNSU patients and 13 unaffected individuals ., The PGE-M levels were measured in duplicate using competitive enzyme-linked immunosorbent assays ( Cayman Chemical , Ann Arbor , MI , USA ) ., We analyzed the exon 7 and exon 10 boundary mutations using RT-PCR to examine the effects of the splice-site mutations on SLCO2A1 transcription ., Total RNA was extracted from biopsy specimens of the small intestine and peripheral blood mononuclear cells using a NucleoSpin RNA Kit ( Macherey-Nagel , Düren , Germany ) or PAXgene Blood RNA Kit ( Qiagen , Hilden , Germany ) ., cDNA was synthesized using a PrimeScript First Strand cDNA Synthesis Kit ( TaKaRa , Otsu , Japan ) ., The PCR products obtained from the cDNAs were sequenced ( S5 Table ) ., A full-length cDNA ( NM_005630 ) expression vector and C-terminally GFP-tagged cDNA expression vector were purchased from OriGene Technologies ( Rockville , MD , USA ) ., To construct vectors carrying a mutated cDNA , a KOD -Plus- Mutagenesis Kit ( Toyobo , Osaka , Japan ) was used according to the manufacturer’s instructions ., The expression vectors were amplified by inverse PCR with specific primer sets ( S6 Table ) ., The PCR products were self-ligated , and transformed into Escherichia coli chemically competent DH5α cells ., To correct a frameshift in the downstream of exon 7 , a C-terminally GFP-tagged cDNA expression vector with deletion of exon 7 was amplified again ., On the day before transfection , HEK293 cells were trypsinized , counted , and plated onto 12-well plates at a density of 4×105 cells/well ., The cells were transfected by adding a premixed solution containing 0 . 4 μg of expression vectors and 2 μl of ScreenFectA ( Wako , Osaka , Japan ) ., After 24 hours of incubation , the medium was exchanged twice with warmed Waymouth’s medium ( Life Technologies , Carlsbad , CA , USA ) , and the cells were incubated for 30 minutes at 37°C in uptake medium containing 5 , 6 , 8 , 11 , 12 , 14 , 15-3H ( N ) -PGE2 ( PerkinElmer , Waltham , MA , USA ) at 0 . 6 nM ., The cells were washed four times with cold Waymouth’s medium , and lysed with 200 μl of RIPA Buffer ( Thermo Fisher Scientific , Hemel Hempstead , UK ) containing a protease inhibitor ( Roche , Basel , Switzerland ) ., The total protein concentration was quantified using a BCA Protein Assay Kit ( Thermo Fisher Scientific ) ., Next , 150 μl of cell lysate was mixed with 5 ml of MicroScint–20 ( PerkinElmer ) , and scintillation counting was performed in a Tri-Carb 3100TR liquid scintillation spectrometer ( PerkinElmer ) ., Formalin-fixed paraffin-embedded tissues were sectioned at 3-μm thickness ., After antigen unmasking in 10 mM sodium citrate buffer ( pH 6 ) for 15 minutes at 121°C , the sections were blocked with Protein Block Serum-Free ( Dako , Glostrup , Denmark ) for 30 minutes at room temperature ., The sections were then incubated with a diluted anti-SLCO2A1 antibody ( HPA013742; Sigma-Aldrich , St . Louis , MO , USA; antigen sequence: PSTSSSIHPQSPACRRDCSCPDSIFHPVCGDNGIEYLSPCHAGCSNINMSSATSKQLIYLNCSCVTGGSASAKTGSCPVPCAH ) overnight at 4°C , followed by MAX-PO ( MULTI ) ( Nichirei , Tokyo , Japan ) for 30 minutes at room temperature ., DAB solution ( Nichirei ) was applied for color development ., After the immunocytochemistry , the sections were counterstained with Mayer’s hematoxylin solution ( Nichirei ) ., For immunofluorescence , the sections were incubated with a primary antibody mixture of the anti-SLCO2A1 antibody ( HPA013742 ) and an anti-VE-cadherin antibody ( LS-B3780; LifeSpan BioSciences , Seattle , WA , USA ) overnight at 4°C , followed by a secondary antibody mixture of Alexa Fluor 568-conjugated goat anti-rabbit IgG ( H&L ) antibody and Alexa Fluor 488-conjugated goat anti-mouse IgG ( H&L ) antibody ( Life Technologies ) for 30 minutes at room temperature ., The stained sections were analyzed using an ECLIPSE TE2000-U ( Nikon , Tokyo , Japan ) ., For observation of HEK293 cells expressing GFP-fusion proteins , cells were fixed with 4% paraformaldehyde phosphate buffer solution ( Wako ) for 20 minutes , and then permeabilized with 0 . 1% Triton X–100 ( Sigma-Aldrich ) in D-PBS ( - ) solution ( Wako ) for 20 minutes ., Nuclei were stained with 16 . 2 μM Hoechst 33342 ( Life Technologies ) in D-PBS ( - ) solution for 5 minutes ., GFP and nuclei were visualized using a 40× objective on an LSM710 Laser Scanning Confocal Microscope ( Carl Zeiss , Oberkochen , Germany ) ., The chi-square test and Fisher’s exact test , where appropriate , were used to analyze categorical data ., Student’s t-test was used to compare quantitative data between two groups ., Dunnett’s method was used for multiple comparisons with a control group ., The analyses were performed using JMP Pro statistical package 11 . 2 . 0 ( SAS Institute , Cary , NC , USA ) ., Values of p < 0 . 05 were regarded as statistically significant .

Introduction, Results, Discussion, Materials and Methods

Previously , we proposed a rare autosomal recessive inherited enteropathy characterized by persistent blood and protein loss from the small intestine as chronic nonspecific multiple ulcers of the small intestine ( CNSU ) ., By whole-exome sequencing in five Japanese patients with CNSU and one unaffected individual , we found four candidate mutations in the SLCO2A1 gene , encoding a prostaglandin transporter ., The pathogenicity of the mutations was supported by segregation analysis and genotyping data in controls ., By Sanger sequencing of the coding regions , 11 of 12 other CNSU patients and 2 of 603 patients with a diagnosis of Crohn’s disease were found to have homozygous or compound heterozygous SLCO2A1 mutations ., In total , we identified recessive SLCO2A1 mutations located at seven sites ., Using RT-PCR , we demonstrated that the identified splice-site mutations altered the RNA splicing , and introduced a premature stop codon ., Tracer prostaglandin E2 uptake analysis showed that the mutant SLCO2A1 protein for each mutation exhibited impaired prostaglandin transport ., Immunohistochemistry and immunofluorescence analyses revealed that SLCO2A1 protein was expressed on the cellular membrane of vascular endothelial cells in the small intestinal mucosa in control subjects , but was not detected in affected individuals ., These findings indicate that loss-of-function mutations in the SLCO2A1 gene encoding a prostaglandin transporter cause the hereditary enteropathy CNSU ., We suggest a more appropriate nomenclature of “chronic enteropathy associated with SLCO2A1 gene” ( CEAS ) .

Advanced diagnostic innovations such as capsule endoscopy and balloon endoscopy have provided better understanding of endoscopic findings of small bowel diseases ., However , it remains difficult to diagnose small intestinal diseases such as Crohn’s disease , intestinal tuberculosis , and nonsteroidal anti-inflammatory drug-induced enteropathy by the endoscopic findings alone ., We previously reported a rare autosomal recessive inherited enteropathy characterized by persistent blood and protein loss from the small intestine ., This enteropathy has an intractable clinical course with ineffectiveness of immunosuppressive treatment ., In this study , we identified recessive mutations in the SLCO2A1 gene , encoding a prostaglandin transporter , as causative variants of this disorder by exome sequencing of four families , and showed that this disease is distinct from Crohn’s disease ., We also showed that the mutations found in the patients caused functional impairment of prostaglandin E2 uptake within cells ., The present findings suggest that genetic analysis together with detailed clinical information is invaluable for diagnosis of the disease , and that there may be a concept of enteropathy referred to as “prostaglandin-associated enteropathy” , irrespective of ethnic background .

journal.pntd.0007757

Intrinsic and extrinsic drivers of transmission dynamics of hemorrhagic fever with renal syndrome caused by Seoul hantavirus

There are 60 , 000–100 , 000 cases of hemorrhagic fever with renal syndrome ( HFRS ) , a rodent-borne zoonosis , reported annually , globally 1 , 2 ., HFRS is caused by subtypes of hantavirus , each of which is associated with a distinct rodent species 3 ., Humans are usually infected by inhaling air contaminated with saliva , feces , or urine from infected rodents 4 ., China is the main epidemic region for HFRS , and accounts for 90% of cases , globally 1 ., The two predominant hantavirus strains circulating in endemic areas of China are Hantaan virus ( HTNV ) , which is carried by striped field mice ( Apodemus agrarius ) , and Seoul virus ( SEOV ) , which is carried by Norway rat ( Rattus norvegicus ) 5 ., HFRS epidemics vary significantly across seasons and are influenced by extrinsic factors across regions , like climate 6–10 ., Following the Hantavirus Pulmonary Syndrome outbreaks that occurred in the Four Corners region of the southwestern United States in 1993 , the correlation between hantavirus infections and climatic conditions was described using a cascade hypothesis 11 ., The hypothesis posited that favorable climate conditions would lead to more available food , and to greater rodent population sizes , thereby enhancing the risk of hantavirus infections ., Similarly , the 2007 HFRS outbreak in temperate southern Europe may have been caused by increased population density of the bank vole , the vector of Puumala virus ( PUUV ) , which , in turn , may have resulted from abundant food due to preceding warmer than usual autumn and winter weather 12 , 13 ., In central China , higher temperature and precipitation in the previous summer led to favorable food conditions for the striped field mouse , the rodent host of HTNV , which led to an autumn peak in incidence of HFRS 14 ., Hantaviruses carried by species of wild rodents , such as HTNV and PUUV , have been extensively studied , but SEOV is vectored by the Norway rat , a domestic rodent that is found in urban environments and is associated with humans 15–17 ., SEOV infections , have milder symptoms and lower fatality rates ( 1–2% vs . 5–10% ) than HTNV infections , so less attention is focused on SEOV 16 , 17 ., Besides , SEOV infections have a higher asymptomatic infection rate of 8–20% , compared with HTNV at 1–4% 18 ., In the past two decades , SEOV-related HFRS cases have been reported in the United Kingdom ( 2012 ) 19 , France ( 2014 ) 20 , the United States , and Canada ( 2017 ) 21 and most provinces of China 22–26 ., It has been estimated that more than 25% of hantavirus infections in South Korea and China are caused by SEOV 27 , 28 ., To better control the potential threat , it is important to understand how critical factors impact SEOV-related HFRS transmission dynamics ., Here , we explore the roles of related intrinsic and extrinsic factors in transmission dynamics of HFRS caused by SEOV in the pre-vaccination era ., Intrinsic factors refer mainly to herd immunity or host population dynamics , and extrinsic factors include external forces such as climatic or environmental factors ., We collected detailed epidemiological and rodent trapping data from a typical SEOV-endemic area of China , Huludao City , where the Norway rat is the dominant rodent species in residential areas 29 ., Additionally , we investigated potential causal relationships among the environmental variables , the population dynamics of the rodent reservoir , and the incidence of HFRS ., We subsequently constructed a mathematical model to quantify intrinsic transmission dynamics and extrinsic effects ., Huludao City ( 40°56′N , 120°28′E ) in the Liaoning Province of China ( S1 Fig ) has a temperate monsoon climate with a hot and rainy summer ( mean temperature , 23 °C; mean monthly precipitation , 123 mm ) and a cold and rainless winter ( mean temperature , -5 . 7 °C; mean monthly precipitation , 2 mm ) ., The city covers 10 , 434 square km and has a population of 2 . 6 million people ., HFRS has always posed a severe threat to human health in Huludao City since 1984 30 ., To control HFRS transmission , mass vaccination campaigns against hantavirus were conducted beginning in 2005 , and combined with rat extermination programs during 2005 and 2012 31 ., Local demographic data were collected from the Liaoning Statistical Yearbooks ., Land cover changes from 1998 to 2015 in Huludao City were derived from the annual European Space Agency ( ESA ) Climate Change Initiative ( CCI ) land cover maps , with a 300 m spatial resolution ., Meteorological data for Huludao City from 1998 to 2015 were obtained from the Chinese Bureau of Meteorology , and then processed into monthly climate data , including monthly mean maximum temperature , mean temperature , mean minimum temperature , cumulative precipitation , relative humidity , and absolute humidity ., Absolute humidity is calculated from temperature and relative humidity 32 ., Data of human HFRS cases in Huludao City from 1998 to 2015 , were obtained from the Huludao City Center for Disease Control and Prevention , China ., All HFRS cases were confirmed according to the standard diagnosis set out by the Ministry of Health of the People’s Republic of China , and by detecting antibodies against hantavirus in serum samples ., Serum samples were tested for immunoglobulin ( Ig ) G and IgM antibodies against HTNV and SEOV by indirect immunofluorescent assay ( IFA ) 33 ., Serum-epidemiological surveys on human hantavirus infection were conducted annually between 1998 and 2015 ., Anonymous ( non-personal ) information was used ., Rodent population density was investigated indoors and outdoors by the powder-trace method on a monthly basis , obtained from the historical literature 34 ., According to the criteria , 400 powdered panels are placed in the special sites for spot checks across the city every month , of which 240 panels are set in restaurant , hotel , station and other public places where rodents are commonly found , and 160 panels are set in general household ., Two powdered panels were placed in each room ( about 15 m2 ) in selected spots ., Outdoors , each panel was placed along a wall at 5–10 m intervals ., All panels were set at night and recovered before sunrise ., Rodent population density was calculated as the number of positive powdered panels ( with footprints or tail tracks from rodents ) divided by the number of effective powdered panels used ., The study protocol was reviewed by the institutional review board of the Huludao Municipal CDC and ethics approval was not required ., We have received consent from home/land owners to collect rodent data on private land and in private homes ., The Animal Ethics Committee of the Huludao Municipal CDC also waived approval for this study ., Wavelet analysis is widely used in ecology and epidemiology studies to explore the variety in the periodicity of a time series through its decomposition properties 35 ., Here , the Morlet wavelet was used to detect the non-stationary characteristics of the incidence of HFRS fluctuations over time , and the bootstrapping method 36 was performed to test the statistical significance of the results ., In the significance test , 1000 surrogate datasets were simulated by bootstrapping to test the null hypothesis , where a P value of < 0 . 05 was considered to be statistically significant ., Convergent cross mapping ( CCM ) is used to detect nonlinear causal relationships between time series of HFRS incidence ( Y ) and environmental factors ( X ) , which is designed to measure causality in nonlinear dynamical systems ., CCM can help to identify bidirectional causality ( i . e . X and Y are mutually coupled ) or unidirectional causality ( e . g . , X time series variable in a system has a causal influence on Y , but not vice versa ) 37 , 38 ., In this study , CCM was used to identify time-delayed effects of a causal interaction between time series of environmental variability , the population dynamics of rodent hosts , and HFRS incidence based on nonlinear state space reconstruction 39 ., In a system where x causes y , Sugihara et al . purposed that the state of x ( t ) can be estimated from the reconstruction of y ( t ) using the nearest-neighbor forecasting method , also called cross mapping 40 ., Pearson’s correlation between the estimated x ( t ) and observed x ( t ) reflects the causal effect of x on y , called “cross map skill” 39 ., In analysis , the embedding dimension ( E ) were set according to the simplex projection results 40 ., Time-delayed effect ( tp ) were set as 0–6 months for detecting the time lags 41 ., Given the seasonality of HFRS incidence , climate variables , and rodent population density , the seasonal component and the response of HFRS risk and rodent population density to the anomalies of other variables were examined to avoid spurious correlations ., For each variable , 1500 surrogate time series having the same degree of shared seasonality , but with randomized anomalies , were generated for seasonal surrogate test ., For a specific variable , the month of year average ( seasonal cycle ) was calculated first , and the seasonal anomaly was represented by the difference between the observed value and the seasonal cycle ., Then the random shuffling of the time series data of seasonal anomalies was added back to the season average 42 ., The above analysis was implemented in the “rEDM” package in R . We estimated the epidemiological parameters for HFRS epidemics in Huludao City by fitting the time series from the observed monthly incidence and rodent population density to a discrete-time susceptible-infection model in the Bayesian framework ( Table 1 ) 43 ., During 1998 to 2015 , 6796 HFRS cases were reported in Huludao City , with an annual incidence from 2 . 50 ( 1/100 , 000 , minimum in 2008 ) to 26 . 99 ( 1/100 , 000 , maximum in 2005 ) ., The pattern of HFRS incidence showed a significant seasonality with an annual peak in spring from March to May ( Fig 1 , S2 Fig ) ., The epidemic variability could be divided into three periods: period I ( 1998–2004 ) , characterized by a high-level of HFRS incidence of over 17 . 6 ( 1/100 , 000 ) and a stable periodicity without limited intervention; period II ( 2005–2012 ) , vaccination campaign and rat extermination program were implemented simultaneously and the local HFRS incidence decreased dramatically to a relatively low level at 10 . 0 ( 1/100 , 000 ) ; period III ( 2013–2015 ) , the incidence of HFRS gradually rebounded ., By using wavelet analysis , we found changes in endemic periodicity after 2006 ( Fig 1B ) , possibly due to the rat extermination program and mass vaccination campaign that were conducted beginning in 2005 ., Here , to reduce the effect of human interventions on disease dynamics , we focused our analysis on the study period between 1998 and 2004 ( period I ) ., In period I , a total of 3914 HFRS cases were reported ., The local rodent community mainly comprised Norway rat and the house mouse , with the former accounting for 94 . 52% of the total ., The population density of the rodents ranged from 4 . 89% to 7 . 28% , with two relatively low values in January of 1998 and 1999 ., We first examined the impact of land use and land cover changes on HFRS transmission ., Land use changed very slightly in period I ( S3 Fig ) , so the effect of habitat and land use change on rodent hosts and disease transmission were minimal during this time period ., We then conducted a cross-correlation analysis of the climate variables , rodent population density , and HFRS incidence ., The Norway rat population was found to be positively correlated with mean temperature ( Pearson’s r = 0 . 23 , P < 0 . 05 ) , maximum temperature ( r = 0 . 25 , P < 0 . 05 ) , minimum temperature ( r = 0 . 23 , P < 0 . 05 ) , cumulative precipitation ( r = 0 . 28 , P < 0 . 05 ) , and absolute humidity ( r = 0 . 24 , P < 0 . 05 ) , with the maximum correlation coefficients occurring at 3 , 3 , 4 , 4 , and 4-month time lags , respectively ( S4A Fig ) ., HFRS incidence was strongly correlated with the climate variables ., All the tested climate variables showed significant negative correlations with HFRS incidence , and maximum cross-correlations between temperature and HFRS incidence occurred at a lag of 3 months and other climate variables at a lag of 2 months ( S4B Fig ) ., Given the survival time of hantavirus outside the host and the incubation period for HFRS 51 , 52 , the longest time lag was set as 6 months ., Convergent cross mapping ( CCM ) was used to detect causality between these time series ., However , rodent data had weak cross mapping skills for the climate variables ., Only a marginally significant , weak causal effect was found between minimum temperature ( with a 4-month lag , tp = -4 ) and absolute humidity ( with a 3-month lag , tp = -3 ) on rodent population density , with cross map skills of 0 . 18 and 0 . 12 , respectively ( S5 Fig ) ., To test the non-stationary ( transient ) correlation between Norway rat and climate variation , we conducted a further test of wavelet coherence analysis ., The results showed scattered and small-sized distribution of significant areas with inconsistent phase differences ( S6 Fig ) ., In all , the combined results indicate that there might be a weak causal relationship between Norway rat and climate variables ., Only relative humidity with a 1-month lag was identified as a significant causal factor for HFRS transmission ( cross map skill , 0 . 86 ) by the CCM method ( S7 Fig ) ., The significance test distinguished anomalies of climate effects from shared seasonal cycle effects ( S7B Fig ) ., To quantify the effect of climatic factors on rodent population density , we tested the models ( see Materials and methods for details ) with θr containing, ( i ) minimum temperature ,, ( ii ) absolute humidity ,, ( iii ) both variables , and, ( iv ) none of these climate variables ., The goodness of fit for all the candidate models is provided in Table 1 ., The model without the climate influence was regarded as optimal to infer the population dynamics of Norway rat in a local urban setting ( DIC = 9 . 17 , R2 = 0 . 50 , Fig 2A ) ., The reproduction rates of the local Norway rat population , as estimated by our model , peaked in June , July , and October , and reached the lowest level in May and August ( Table 2 ) ., The average life span of the Norway rat is estimated to be about 6 months , and the local environmental carrying capacity is estimated to be 8 . 24 ., The trace plot of Markov chains and the posterior distribution for the parameters in the optimal rodent model is shown in S8 Fig . All the parameter traces have passed the convergence test ., Based on the causal relationship tested by CCM , we constructed a HFRS dynamic model containing the infected rodent population and seasonal contact rate to quantify the specific effect of relative humidity on risk of HFRS ( see Materials and methods for details ) ., We constructed two models: one containing relative humidity and the other without ., The model with relative humidity had a slightly better fit when the simulated HFRS cases correlated well with the observations and 75% of the variance was explained ( R2 = 0 . 75 , DIC = 5 . 57 ) ( Fig 2B ) , relative to the model without relative humidity ( R2 = 0 . 72 , DIC = 7 . 89 ) ., The optimal models also captured the main seasonal patterns of both the rodent population density and reported cases ( Fig 2C and 2D ) ., The parameter γ in βcli = δ3RH γ t-1 , was estimated to be −1 . 23 ( 95% CI , -1 . 62– -0 . 83 ) ( Fig 3A ) , indicating that relative humidity had a negative impact on HFRS transmission ( Fig 3B ) ., Additionally , during winter–spring , the estimated seasonal contact rates peaked from February to April ( Fig 4 ) , which may result from the behaviors of human and rodent ., Most of the patients are farmers ( 77% ) and it’s the slack time of winter-spring when people stay at home ., Besides , due to the severe weather outside during that time , Norway rats would live and feed in closer proximity to human residence for favorable living conditions and available food , which could be a reason for the high contact rate ., The trace plot of Markov chains and the posterior distribution for the parameters used in the optimal HFRS transmission model are shown in S9 Fig ., We used a mathematical model to quantify the impact of intrinsic and extrinsic factors on risk of HFRS , and to test the mechanistic understanding of HFRS transmission dynamics in a SEOV endemic area of China ., By analyzing time-series data of Norway rat populations and SEOV infections , we found that the population dynamics of Norway rat in urban settings were relatively well-predicted by a simple model which excluded climatic conditions ., Unusually , we demonstrated that a potential biotic driver ( relative humidity ) could enhance predictive ability of the HFRS transmission analysis ., Our results highlight the crucial role of mass immunization campaigns and rat extermination program in HFRS transmission control ., It should be noted that SEOV infections began to rise again after 2013 in the post-vaccination era ., In HFRS endemic areas where hosts are mainly wild rodents , fluctuations in the population density of wild rodents , such as striped field mouse or bank vole , are influenced by climatic factors and can mediate the effect of climate on the risk of HFRS transmission , mainly because of the close association between climate and host food supply 53 ., However , in the SEOV-endemic area , the CCM results showed that the change in population density of the Norway rat was insensitive to climate variability ., Our results differed from those of a previous study which found a relationship among the virus-carrying index , climate variables , and HFRS incidence by using structural equation modeling ( SEM ) ., This study reported that the Norway rat served as an important mediator of disease transmission when climate variability was found to influence the risk of HFRS 54 ., This inconsistency may be due to a difference in sampling frequency as we surveyed on a monthly basis , while Guan et al . ( 2009 ) sampled quarterly ., Additionally , the SEM analysis used a latent variable to account for measurement error but did not evaluate the contribution of specific variables to the transmission dynamics ., Norway rat is a domestic rodent whose lifestyle differs greatly from those of wild rodents ., Norway rat lives in residential areas and feeds on various food items from humans instead of field crops 55 ., Food resources and habitats are relatively stable and change only slightly with climate variability ., Additionally , Norway rats breed throughout the year in Huludao City , where the winter temperatures rarely drop below −10 °C and residential areas have warmer conditions because of the winter heating policy in China ., Winter breeding of wild rodents is limited because of low temperatures 56 ., Our findings indicated that an abiotic factor , relative humidity , was a critical indicator of HFRS risk in the SEOV-endemic area of Huludao City , which is consistent with previous studies in which SEOV was the main virus type 57 , 58 ., We assumed that relative humidity may be associated with the survival or infectivity of SEOV in the environment , the activity of the rodent , and human hosts or the transmission process , although the underlying mechanism is not clear ., We do not know of any relevant experiments on how environmental conditions affect survival of SEOV outside the host , while PUUV and HTNV have been confirmed to have longer infective periods at low temperatures and high humidities in experimental environments 59 ., In our model , interannual variation in SEOV infections was partly explained by changes in relative humidity ., A less intense outbreak in 2000 might be associated with a lower relative humidity-driven transmission potential ( Fig 2B ) ., Some limitations in our study should be mentioned ., First , rodent population density was extremely low in the springs of 1998 and 1999 , which reduced the overall explanatory ability of our model ., Second , prevalence of SEOV infection in Norway rats was assessed by mathematical model due to lack of available data ., Third , to ameliorate the influence of vaccination campaign and other interventions , only the selected time range of data was used to detect the relationship between intrinsic/extrinsic drivers and SEOV transmission ., Further efforts should be made to improve model reliability through efforts such as addition of data to the available dataset ., More attention should be paid to the association between economic development , such as infrastructure improvement , the spatial distribution of Norway rat , and induced SEOV infections across districts in subsequent studies ., The SEOV-related HFRS is posing a health threat to an increasing number of people ., Although the Norway rat is found everywhere , SEOV cases are not ., We suspect this may mainly be due to increased surveillance and attention focused on this pathogen , as SEOV previously caused many asymptomatic human infections previously ., In conclusion , based on the longitudinal and complete dataset , our study yielded a proper framework for understanding the intrinsic transmission dynamics and extrinsic effects on HFRS risk caused by SEOV , especially at the human-animal-environment interface ., The framework can be flexibly adjusted when more information is available ., Furthermore , we provided clues to potential environmental drivers on SEOV transmission dynamics , which would be useful for further research related to public health issues .

Introduction, Materials and methods, Results, Discussion

Seoul hantavirus ( SEOV ) has recently raised concern by causing geographic range expansion of hemorrhagic fever with renal syndrome ( HFRS ) ., SEOV infections in humans are significantly underestimated worldwide and epidemic dynamics of SEOV-related HFRS are poorly understood because of a lack of field data and empirically validated models ., Here , we use mathematical models to examine both intrinsic and extrinsic drivers of disease transmission from animal ( the Norway rat ) to humans in a SEOV-endemic area in China ., We found that rat eradication schemes and vaccination campaigns , but below the local elimination threshold , could diminish the amplitude of the HFRS epidemic but did not modify its seasonality ., Models demonstrate population dynamics of the rodent host were insensitive to climate variations in urban settings , while relative humidity had a negative effect on the seasonality in transmission ., Our study contributes to a better understanding of the epidemiology of SEOV-related HFRS , demonstrates asynchronies between rodent population dynamics and transmission rate , and identifies potential drivers of the SEOV seasonality .

Seoul hantavirus ( SEOV ) infections are common in Europe and Asia where a considerably high seroprevalence among the population is found ., However , only relatively few hemorrhagic fever with renal syndrome ( HFRS ) cases are reported ., Comprehensive epidemiological data is necessary to study the patterns and drivers of this underestimated disease ., Here , we analyzed rodent host surveillance and seroprevalence data from 1998 to 2015 for disease outbreaks in Huludao City , one of the typical SEOV-endemic areas for HFRS in China ., Our mathematical models quantified the drivers on HFRS transmission and estimated the epidemiological parameters ., Our study provides an understanding of its ecological process between intrinsic and extrinsic factors , human-rodent interface and disease dynamics .

medicine and health sciences, pathology and laboratory medicine, atmospheric science, population dynamics, pathogens, microbiology, vertebrates, animals, mammals, animal models, viruses, seasons, hemorrhagic fever with renal syndrome, model organisms, hantavirus, rna viruses, experimental organism systems, population biology, humidity, bunyaviruses, research and analysis methods, infectious diseases, animal studies, medical microbiology, microbial pathogens, population metrics, rodents, eukaryota, meteorology, earth sciences, viral pathogens, biology and life sciences, viral diseases, population density, amniotes, organisms, rats

journal.pbio.2002128

Evolutionary restoration of fertility in an interspecies hybrid yeast, by whole-genome duplication after a failed mating-type switch

A whole-genome duplication ( WGD ) occurred more than 100 million years ago in the common ancestor of 6 yeast genera in the ascomycete family Saccharomycetaceae , including Saccharomyces 1 , 2 ., Recent phylogenomic analysis has shown that the WGD was an allopolyploidization—that is , a hybridization between 2 different parental lineages 3 ., One of these parental lineages was most closely related to a clade containing Zygosaccharomyces and Torulaspora ( ZT ) , whereas the other was closer to a clade containing Kluyveromyces , Lachancea , and Eremothecium ( KLE ) ., The ZT and KLE clades are the 2 major groups of non-WGD species in family Saccharomycetaceae ., The WGD had a profound effect on the genome , proteome , physiology , and cell biology of the yeasts that are descended from it , but the genomes of these yeasts have changed substantially in the time since the WGD occurred , with extensive chromosomal rearrangement , deletion of duplicate gene copies , and sequence divergence between ohnologs ( pairs of paralogous genes produced by the WGD ) ., These changes have made it difficult to ascertain the molecular details of how the WGD occurred ., Ancient hybridizations are rare in fungi ( or at least difficult to detect 4 ) , but numerous relatively recent hybridizations have been identified using genomics , particularly in the ascomycete genera Saccharomyces 5 , 6 , Zygosaccharomyces 7–9 , Candida 10–12 , and Millerozyma 13 ., Marcet-Houben and Gabaldón 3 proposed 2 alternative hypotheses for the mechanism of interspecies hybridization that led to the ancient WGD in the Saccharomyces lineage ., Hypothesis A was hybridization between diploid cells from the 2 parental species , perhaps by cell fusion ., Hypothesis B was mating between haploid cells from the 2 parental species to produce an interspecies hybrid zygote , followed by genome doubling ., Under both hypotheses , the product is a cell with 2 identical copies of each parental chromosome ., These identical copies should be able to pair during meiosis , leading to viable spores ., While there are no known examples of natural yeast hybrid species formed by diploid–diploid fusion ( hypothesis A ) , 3 examples have been discovered in which hybrid species were apparently formed simply by mating between haploids of opposite mating types from different species ( hypothesis B ) ., These are Candida metapsilosis 11 , C . orthopsilosis 10 , 12 , and Zygosaccharomyces strain ATCC42981 8 , 14 ., These interspecies hybridizations occurred by mating between parents with 4%–15% nucleotide sequence divergence between their genomes ., However , none of these 3 hybrids can sporulate , which could be either because the homeologous chromosomes from the 2 parents are too divergent in sequence to pair up during meiosis or because pairing occurs but evolutionary rearrangements ( such as translocations ) between the parental karyotypes result in DNA duplications or deficiencies after meiosis 15–18 ., None of these 3 hybrids has undergone the genome-doubling step envisaged in hypothesis B . Several groups 3 , 18–20 have proposed that genome doubling could occur quite simply by means of damage to 1 copy of the MAT locus in the interspecies hybrid , which could cause the hybrid cell to behave as a haploid , switch mating type , and hence autodiploidize ., This proposal mimics laboratory experiments carried out by Greig et al . 21 in which hybrids between different species of Saccharomyces were constructed by mating ., The hybrids were unable to segregate chromosomes properly and were sterile , but when 1 allele of the MAT locus was deleted , they spontaneously autodiploidized by mating-type switching and were then able to complete meiosis and produce spores with high viability ., Each spore contained a full set of chromosomes from both parental species 21 ., While genome doubling via MAT locus damage is an attractive hypothesis consistent with hypothesis B above 3 , no examples of it occurring in nature have been described ., We show here that Z . parabailii has gone through this process ., There are 12 formally described species in the genus Zygosaccharomyces 22 ., The most studied of these is Z . rouxii , originally found in soy sauce and miso paste 23 , 24 ., Others include Z . mellis , frequently found in honey 25 , and Z . sapae from balsamic vinegar 26 , 27 ., Species in the Z . bailii sensu lato clade ( Z . bailii , Z . parabailii , and Z . pseudobailii; 28 ) are of economic importance because they are exceptionally resistant to osmotic stress and low pH . Their resistance to the weak organic acids commonly used as food preservatives makes them the most frequent spoilage agent of packaged foods with high sugar content , such as fruit juices and jams , or with low pH , such as mayonnaise 29–33 ., These same characteristics make Zygosaccharomyces relevant to biotechnology since high stress tolerance and rapid growth are often desirable traits in microorganisms to be used as cell factories ., The strain we analyze here , Z . parabailii ATCC60483 , has previously been used for production of vitamin C 34 , lactic acid 35 , and heterologous proteins 36 ., Despite the diversity of the genus , genome sequences have been published for only 2 nonhybrid species of Zygosaccharomyces: the type strains of Z . rouxii ( CBS732T; 37 ) and Z . bailii ( CLIB213T; 38 ) ., The genus also includes many interspecies hybrids with approximately twice the DNA content of pure species ( 20 Mb instead of 10 Mb; 7 , 8 , 14 , 39 ) ., Mira et al . 39 sequenced the genome of Zygosaccharomyces strain ISA1307 and found that it is a hybrid between Z . bailii and an unidentified Zygosaccharomyces species ., In 2013 , Suh et al . 28 proposed that some strains that were historically classified as Z . bailii should be reclassified as 2 new species , Z . parabailii and Z . pseudobailii , based on phylogenetic analysis of a small number of genes ., The sequences of the RPB1 and RPB2 genes that they obtained from Z . parabailii and Z . pseudobailii contained multiple ambiguous bases , consistent with a hybrid nature 39 ., In the current study , we sequenced the genome of a second hybrid strain , ATCC60483 ., We show that ATCC60483 and ISA1307 are both Z . parabailii and are both descended from the same interspecies hybridization event ., By sequencing ATCC60483 using Pacific Biosciences ( PacBio ) technology , we obtained near-complete sequences of every Z . parabailii chromosome , which enabled us to study aspects of chromosome evolution in this species that were not evident from the Illumina assembly of ISA1307 39 ., We first tried to sequence the Z . parabailii genome using Illumina technology , but even with high coverage , we were unable to obtain long contigs ., The data indicated that the genome was a hybrid , so instead we switched to PacBio technology , which generates long sequence reads ( 6 kb on average in our data ) ., Our initial assembly had 22 nuclear scaffolds , which we refined into 16 complete chromosome sequences with a cumulative size of 20 . 8 Mb by manually identifying overlaps between the ends of scaffolds and by tracking centromere and telomere locations ., We annotated genes using the Yeast Genome Annotation Pipeline ( YGAP ) , assisted by RNA sequencing ( RNA-Seq ) data to identify introns ., The nuclear genome has 10 , 087 protein-coding genes , almost twice as many as Z . bailii CLIB213T ( Table 1 ) ., Most of the chromosome sequences extend into telomeric repeats at the ends ., The consensus sequence of the telomeres is tgtgggtgggg , which matches exactly the sequence of the template region of the 2 homeologous TLC1 genes for the RNA component of telomerase that are present in the genome ., Chromosome sequences that do not extend into telomeres instead terminate at gene families that are amplified in subtelomeric regions or contain genes that are at chromosome ends in the inferred Ancestral ( pre-WGD ) gene order for yeasts 41 indicating that they are almost full length , except for 3 chromosome ends that appear to have undergone break-induced replication ( BIR ) and homogenization with other chromosome ends ., We identified 1 scaffold as the mitochondrial genome , which maps as a 30-kb circle containing orthologs of all S . cerevisiae mitochondrial genes ., We also found a plasmid in the 2-micron family ( 5 , 427 bp ) , with 99% sequence identity to pSB2 , which was first isolated 42 from the type strain of Z . parabailii ( NBRC1047/ATCC56075 ) ., Visualization of the genome using a Circos plot 43 shows that most of the genome is duplicated , indicating a polyploid origin ( Fig 1 ) ., However , although most genes have a homeolog , the chromosomes do not form simple collinear pairs ., Instead , sections of each chromosome are collinear with sections of other chromosomes ., Comparison to Z . bailii CLIB213T shows that for each region of the Z . bailii genome , there are 2 corresponding regions of the Z . parabailii genome: 1 almost identical in sequence and 1 with approximately 93% sequence identity , which demonstrates a hybrid ( allopolyploid ) origin of Z . parabailii and suggests that Z . bailii was one of its parents ., To analyze this relationship in detail , we estimated the parental origin of every Z . parabailii ATCC60483 gene based on the number of synonymous substitutions per synonymous site ( KS ) when compared to its closest Z . bailii homolog ( Fig 2A ) ., This analysis revealed a bimodal distribution of KS values in which 47 . 1% of the ATCC60483 genes are almost identical to CLIB213T genes ( KS ≤ 0 . 05 ) and a further 42 . 5% are more divergent ( 0 . 05 < KS ≤ 0 . 25 ) ., From this relationship , we infer that Z . parabailii ATCC60483 is an interspecies hybrid formed by a fusion of 2 parental cells , which we refer to as Parent A ( purple ) and Parent B ( green ) ., Parent A was a cell with a genome essentially identical to Z . bailii CLIB213T ., Parent B was a cell of an unidentified Zygosaccharomyces species with approximately 93% overall genome sequence identity to Z . bailii , corresponding to a synonymous site divergence peak of KS = 0 . 16 ( Fig 2A ) ., We refer to the 2 sets of DNA in Z . parabailii that were derived from Parents A and B as the A-subgenome and the B-subgenome , respectively ., We refer to the A- and B-copies of a gene as homeologs , and we use a suffix ( “_A” or “_B” ) in gene names to indicate which subgenome they come from ., The genome contains ribosomal DNA ( rDNA ) loci inherited from each of its parents ., Our assembly includes 2 complete rDNA units with 26S , 5 . 8S , 18S , and 5S genes ., Phylogenetic analysis of their internal transcribed spacer ( ITS ) sequences shows that the rDNA on chromosome 11 is derived from Z . bailii ( Parent A ) , whereas the rDNA on chromosome 4 is derived from Parent B and contains an ITS variant seen only in other Z . parabailii strains ( S1 Fig ) ., A third rDNA locus in our assembly ( at 1 telomere of chromosome 15 ) is incomplete and does not extend into the ITS region ., The rDNA unit on chromosome 4 is also telomeric , whereas the unit on chromosome 11 is located at an internal site 165 kb from the right end ., None of the genes in the interval between this rDNA and the right telomere of chromosome 11 have orthologs in Z . bailii CLIB213T ., Z . parabailii has 16 chromosomes ., We identified its 16 centromeres bioinformatically , which correspond to 2 copies ( A and B ) of each of the 8 centromeres in the Ancestral pre-WGD yeast genome ( Table 2 ) 41 , 44 ., In contrast , Z . rouxii has only 7 chromosomes because of a telomere-to-telomere fusion between 2 chromosomes followed by loss of a centromere 44 ., The missing centromere in Z . rouxii is Ancestral centromere Anc_CEN2 , which maps to Z . parabailii centromeres CEN4 and CEN11 , located between the genes MET14 and VPS1 ., The Z . rouxii centromere must have been lost after it diverged from the Z . bailii/Z ., parabailii lineage ., Alignment of the Z . rouxii MET14-VPS1 intergenic region with the Z . parabailii CEN4 and CEN11 regions shows that the CDE III motif of the point centromere has been deleted in Z . rouxii ( S2 Fig ) ., Z . parabailii inherited the mitochondrial genome of its Z . bailii parent ., A complete mitochondrial genome sequence for Z . bailii is not available , but we identified 55 small mitochondrial DNA ( mtDNA ) contigs in the CLIB213T assembly , which together account for most of the genome , and calculated an average of 96% sequence identity between these and ATCC60483 mtDNA ., CLIB213T lacks 2 of the 5 mitochondrial introns that are present in ATCC60483: the omega intron of the large subunit mitochondrial rDNA and intron 2 of COX1 ., Intraspecies polymorphism for intron presence/absence and comparable levels of intraspecies mtDNA sequence diversity have been reported in other yeast species 45 , 46 ., When genes in the Circos plot are colored according to their parent of origin , it is striking that many Z . parabailii chromosomes are either almost completely “A” ( purple ) or almost completely “B” ( green ) ( outer ring in Fig 1 ) , even though the chromosomes do not form collinear pairs ., This pattern can be seen in more detail in a dot-matrix plot between Z . bailii and Z . parabailii ( Fig 3 ) ., From this plot , it is evident that most of the A-subgenome is collinear with Z . bailii scaffolds , whereas the B-subgenome contains many rearrangements relative to Z . bailii ., For example , Z . parabailii chromosome 1 is derived almost entirely from the B-subgenome but maps to about 12 different regions on the Z . bailii scaffolds ., In contrast , Z . parabailii chromosome 3 is derived from the A-subgenome and is collinear with a single Z . bailii scaffold ., In total , from Fig 3 we estimate that there are approximately 34 breakpoints in synteny between the Z . parabailii B-subgenome and Z . bailii but no breakpoints between the A-subgenome and Z . bailii , when posthybridization rearrangement events ( described below ) are excluded ., This difference in the levels of rearrangement in the A- and B-subgenomes relative to Z . bailii indicates that the 2 subgenomes were not collinear at the time the hybrid was formed ., Therefore , most of the rearrangements between the 2 subgenomes are rearrangements that existed between the 2 parental species prior to hybridization ., The 2 parents both had 8 chromosomes , but their karyotypes were quite different ., Because each event of reciprocal translocation or inversion creates 2 synteny breakpoints 47 , we estimate that about 17 events of chromosomal translocation or inversion occurred between the 2 parents in the time interval between when they last shared a common ancestor and when they hybridized ., The situation in Z . parabailii ( hybridization between parents differing by 17 rearrangements and 7% sequence divergence ) contrasts with that in the hybrid Millerozyma sorbitophila ( only 1 detectable rearrangement between the parents , despite 15% sequence divergence 13 ) ., Although the Z . parabailii genome largely contains unrearranged parental chromosomes , there have been 2 major types of rearrangement after hybridization ., First , posthybridization recombination between the subgenomes at homeologous sites has formed some chromosomes that are partly “A” and partly “B . ”, Second , a process of homogenization has occurred at some places in which 1 subgenome overwrote the other , resulting in gene pairs that are A:A or B:B ., This process is commonly called loss of heterozygosity ( LOH ) or gene conversion ., Based on their KS distances from Z . bailii , the genome contains 4 , 153 simple A:B homeologous gene pairs , 300 A:A pairs , and 84 B:B pairs ., To examine the genomic locations of LOH and rearrangement events in more detail , we further classified genes using a scheme that takes account of their pairing status as well as their divergence from Z . bailii ., Genes were defined as “A” or “B” as before or “N” if a KS distance from Z . bailii could not be calculated ( Fig 2B and 2C ) ., We then assigned each gene to 1 of 7 categories such as “B-gene in an A:B pair” or “A-gene , unpaired” and plotted the locations of genes in each category ., The resulting map of the genome ( Fig 4 ) allows LOH and recombination events to be visualized ., N-genes ( black in Fig 4 ) are seen to be mostly located near telomeres ., Several points of recombination between the A- and B-subgenomes are apparent , such as in the middle of chromosome 4 ., LOH tends to occur in stretches that span multiple genes ., For example , on chromosome 13 , LOH has formed 8 runs of consecutive A-genes in a chromosome that is otherwise “B”; these A-genes are members of A:A pairs ., They were probably formed by homogenization ( gene conversion without crossover ) , although they could also be the result of double crossovers followed by meiotic segregation of chromosomes ., Patches of LOH are frequently seen adjacent to sites of recombination between the 2 subgenomes ( Fig 4 ) ., Three large regions of apparently unpaired A-genes near the ends of chromosomes ( 1L , 5L , and 9R; light blue in Fig 4 ) are probably artefacts caused by BIR , which is a process that can make the ends of 2 chromosomes completely identical from an initiation point out to the telomere 48 ., These regions have 2x sequence coverage in our Illumina data , and we can identify the probable locations of an identical second copy of each of them at other chromosome ends ( Fig 4 ) ., The Z . parabailii genome contains 2 MAT loci ( one of which is broken ) and 4 HML/HMR silent loci ( Fig 5 ) ., In S . cerevisiae , mating-type switching is a DNA rearrangement process that occurs in haploid cells to change the genotype of the MAT locus 49 ., During switching , the active MAT locus is first cleaved by an endonuclease called HO , and its a- or α-specific DNA is removed by an exonuclease ., The resulting double-strand DNA break at MAT is then repaired by copying the sequence of either the HMLα or HMRa locus ., This process converts a MATa genotype to MATα , or vice versa ., Repeated sequences , called Z and X , located beside MAT and the HM loci act as guides for the DNA strand exchanges that occur during this repair process ., The HM loci are “silent” storage sites for the a and α sequence information because genes at these loci are not transcribed due to chromatin modification; only MAT is transcribed 49 ., We infer that the parents of Z . parabailii each contained a MAT locus and 2 silent loci ( HMLα and HMRa ) , similar to S . cerevisiae and Z . rouxii haploids 50 ., Fig 5A shows that Z . parabailii has a MAT locus on chromosome 7 , flanked by Z and X repeats and full-length copies of the genes SLA2 and DIC1 , similar to the MAT loci of many other species 50 , 51 ., This MAT locus is derived from Parent A . Chromosome 7 also contains HMLα and HMRa loci ( derived from Parent B ) near its telomeres ., However , the B-subgenome’s MAT locus is broken into 2 pieces ., Most of it is on chromosome 2 , but its left part ( the 3′ end of MATα1 , the Z repeat , and the neighboring gene SLA2 ) is on chromosome 16 ( Fig 5A ) ., Chromosomes 2 and 16 also each contain an HMLα or HMRa locus from the A-subgenome ., Examination of the breakpoint in the B-subgenome’s MAT locus shows that the break was catalyzed by HO endonuclease , because it occurs precisely at the cleavage site for this enzyme ( Fig 5B ) ., In S . cerevisiae , HO has a long ( approximately 18 bp ) recognition sequence that is unique in the genome , and it cleaves DNA at a site within this sequence , leaving a 4-nucleotide 3′ overhang 52 ., Although the recognition and cleavage sites of HO endonucleases in other species have not been investigated biochemically , they can be deduced because the core of the HO cleavage site ( cgcagca ) invariably forms the first nucleotides of the Z region in each species 51 ., Moreover , the HO cleavage site corresponds to an amino acid sequence motif ( faqq ) in the MATα1 protein that is strongly conserved among species ., The 2 parts of the broken MAT locus are located beside the genes GDA1 and YEF1 ( Fig 5A ) , which are neighbors in Z . bailii CLIB213T and in the Ancestral yeast genome 38 , 41 ., Therefore , after HO endonuclease cleaved the “B” MAT locus , the broken ends of the chromosome apparently interacted with the GDA1-YEF1 intergenic region of the A-subgenome , causing a reciprocal translocation ., This site is the only synteny breakpoint between the A-subgenome of Z . parabailii and the genome of Z . bailii ( scaffold 9; Fig 3 ) ., Comparison of the DNA sequences at the site ( Fig 5B ) shows no microhomology between the 2 interacting sequences and that DNA repair led to duplications of a 5-bp sequence ( acaac ) from the GDA1-YEF1 intergenic region and a 2-bp sequence ( ca ) from MATα1 , suggestive of nonhomologous end joining ( NHEJ ) as the repair mechanism ., We hypothesize that this genomic rearrangement occurred during a failed attempt to switch mating types , which resulted in a reciprocal translocation instead of normal repair of MAT by HML or HMR ., While the B-subgenome’s MATα1 gene is clearly broken , its MATα2 gene also appears to be nonfunctional ., MATα2 has 2 introns , and our RNA-Seq data show how both homeologs of this gene ( ZPAR0G01480_A and ZPAR0B05090_B ) are spliced ., A point mutation at the 3′ end of intron 2 of the B-gene changed its AG splice acceptor site to AC , with the result that splicing now uses another AG site 2 nucleotides downstream ( Fig 5C ) ., This change results in a frameshift , truncating the B-copy of the α2 protein to 57 amino acid residues instead of 211 and presumably inactivating it ., Surprisingly , the Z . parabailii genome does not contain any MATa1 ( or HMRa1 ) gene ., This gene codes for the a1 protein , which is 1 subunit of the heterodimeric a1-α2 transcriptional repressor that is formed in diploid ( a/α ) cells and which acts as a sensor of diploidy by repressing transcription of haploid functions such as mating while permitting diploid functions such as meiosis 53 ., The a1 gene is present in Z . rouxii and Z . sapae 27 , 37 , 50 , but it is also absent from Z . bailii CLIB213T and must have been absent from Parent B . The Z . bailii CLIB213T MAT organization is not fully resolved 38 , but it contains a MAT locus with α1 and α2 genes on scaffold 14 and an HMR locus with only an a2 gene on scaffold 19 ., Evolutionary losses of MATa1 have previously been seen in some Candida species 54 , 55 , but not in any species of family Saccharomycetaceae ., In contrast , the gene for the other subunit of the heterodimer , MATα2 , is present in all Zygosaccharomyces species and is probably maintained because it has a second role in repressing a-specific genes in this genus 56 ., Solieri and colleagues have reported evidence that a1-α2 is nonfunctional in a Z . rouxii/pseudorouxii hybrid in which its 2 subunits are derived from different species 14 ., The 2 subgenomes apparent in the Illumina scaffolds of the Zygosaccharomyces hybrid strain ISA1307 , previously sequenced by Mira et al . 39 , are both 99%–100% identical in sequence to the A- or B-subgenomes of ATCC60483 ., Therefore , ISA1307 is also a strain of Z . parabailii ., Importantly , the ISA1307 genome sequence contains the same HO-catalyzed reciprocal translocation between MATα1 of the B-subgenome and the GDA1-YEF1 intergenic region of the A-subgenome ( Fig 5A ) ., Because this rearrangement is so unusual and because it did not involve recombination between repeated sequences , it is highly unlikely to have occurred twice in parallel ., The rearrangement is much more likely to have occurred only once , in a common ancestor of the 2 Z . parabailii strains after the hybrid was formed ., It cannot pre-date the hybridization because it formed junctions between the A- and B-subgenomes , which originated from different parents ., ATCC60483 and ISA1307 are independent isolates of Z . parabailii , both from industrial sources ., ATCC60483 was isolated from citrus concentrate used for soft drink manufacturing in the Netherlands 57 , 58 , and ISA1307 was a contaminant in a sparkling wine factory in Portugal 39 , 59–61 ., We found several examples in which the 2 strains differ in their patterns of LOH , which confirms that they have had some extent of independent evolution ., All 3 large regions of BIR ( on chromosomes 1 , 5 , and 9; Fig 4 ) are unique to ATCC60483 ., ISA1307 contains A:B homeolog pairs throughout these regions , whereas ATCC60483 has only A-genes , which we infer to be in A:A pairs ., Other examples of differential LOH include a 4-kb region around homologs of the S . cerevisiae gene YLR049C , which exists as B:B pairs in ATCC60483 but A:B pairs in ISA1307 , and the gene KAR4 , which is an A:B pair in ATCC60483 but only a B-gene ( single contig ) in ISA1307 ., Notably , the section of the RPB1 gene ( also called RPO21 ) that Suh et al . 28 used for taxonomic identification of Z . parabailii and Z . pseudobailii exists as an A:B pair in ATCC60483 , but only as an A-gene in the ISA1307 genome ., The absence of the B-copy of RPB1 made Mira et al . 39 hesitant to conclude that ISA1307 is Z . parabailii ., Both ATCC60483 and the type strain of Z . parabailii ATCC56075T have previously been reported to be capable of forming ascospores 28 , 57 , 58 ., We confirmed that our stock of ATCC60483 is able to sporulate ( Fig 6A and 6B ) ., On malt extract agar plates , we observed that sporulation occurs directly in zygotes formed by conjugation between 2 cells , resulting in asci in which the 2 former parental cell bodies typically contain 2 ascospores each ., Such dumbbell-shaped ( conjugated ) asci , indicative of sporulation immediately after mating , are characteristic of the genus Zygosaccharomyces 25 and have previously been described in other Z . bailii ( sensu lato ) strains 25 , 62–66 ., The presence of conjugating cells in a culture grown from a single strain indicates that ATCC60483 is functionally haploid ( capable of mating ) and that it is homothallic ( capable of mating-type switching ) ., Since the zygote proceeds immediately into sporulation without further vegetative cell divisions , the diploid state of Z . parabailii appears to be unstable ., Although Suh et al . 28 reported that asci of the type strain of Z . parabailii contain 2 spores , we consistently observed that asci occur in pairs of mated cells connected by a conjugation tube ( Fig 6A and 6B ) , indicating that 4 spores are formed per meiosis ., We dissected tetrad asci from ATCC60483 , grew colonies from the spores , and then used colony PCR to determine their genotype at the intact MAT locus on chromosome 7 ., Among 13 tetrads analyzed , 9 showed a ratio of 2 MATa colonies to 2 MATα colonies ( Fig 6C and 6D ) ., Two tetrads showed 1:3 or 3:1 ratios , and the other two yielded both MATa and MATα PCR products from some single-spore colonies ., The genotype of the ATCC60483 starting strain is MATα from the A-subgenome ( designated MATα_A ) , so the presence of MATa genotypes in colonies derived from spores made by this strain confirms that mating-type switching occurred at some point ., We sequenced the PCR products and found that the A- and B-subgenome HMRa loci were both used as donors for mating-type switching: among the pure MATa colonies , 18 were MATa_A , and 7 were MATa_B ( Fig 6D ) ., Quite surprisingly , 4 tetrads with 2a:2α segregation had 1 MATa_A and 1 MATa_B spore colony , which is inconsistent with simple meiotic segregation from an a/α diploid ., Because all the spores contain a functional HO gene , the genotypes of these 4 tetrads ( #1 , #7 , #19 , and #20 ) probably result from additional switches during the early growth of some colonies ., Similarly , switching during early colony growth may explain the presence of MATα_B genotypes in tetrad #11 and the colonies with mixed a+α genotypes ( in tetrads #11 and #13 ) , as well as the presence of faint PCR products corresponding to the alternative MAT genotype in some other colonies ( Fig 6C ) ., In S . cerevisiae , homothallic diploid ( HO/HO MATa/MATα ) strains show 2:2 segregation of MAT alleles in tetrads , but after spore germination the haploid cells can then switch mating types as often as once per cell division 67 , leading to mating and colonies that contain mostly diploid cells 68; by contrast , most ( but not all ) of the Z . parabailii spore-derived colonies contained a single mating type ( Fig 6C and 6D ) ., We found that almost all the genes involved in mating and meiosis that Mira et al . 39 reported to be missing from the Z . parabailii ISA1307 genome are in fact present in both ATCC60483 and ISA1307 ( S1 Table ) ., For example , we annotated A- and B-homeologs of IME1 , UME6 , DON1 , SPO21 , SPO74 , REC104 , and DIG1/DIG2 as well as MATa2 , MATα1 , and MATα2 ., We also identified genes for the α-factor and a-factor pheromones ( MFα and MFa ) ., The MFα genes code for an unusually high number of copies ( 10–14 ) of a 13-residue peptide whose consensus sequence , ahlvrlspgaamf , is quite different from that of other yeasts , including Z . rouxii ( 7/13 matches ) and S . cerevisiae ( 4/13 matches ) 2 ., Z . parabailii and Z . bailii do lack most of the ZMM group of genes , involved in crossover interference during recombination 69 , even though these are present in Z . rouxii ( S1 Table ) ., Interestingly , identical sets of ZMM genes have been lost in Z . bailii/Z ., parabailii relative to Z . rouxii , as were lost in most Lachancea species relative to Lachancea kluyveri 70: ZIP2 , CST9 ( ZIP3 ) , SPO22 ( ZIP4 ) , MSH4 , MSH5 , and SPO16 are absent , as well as MLH2 , which is not known to be a ZMM gene , whereas ZIP1 is retained ., A similar loss of ZMM genes has occurred in Eremothecium gossypii relative to E . cymbalariae 71 ., A small number of Z . parabailii ATCC60483 genes have “disabling” mutations—frameshifts or premature stop codons that prevent translation of a normal protein product ., The majority of these mutations are present in only 1 subgenome of ATCC60483 and are unique to this strain ., For example , there is a 1-bp insert in the A-homeolog of the DNA repair gene MLH1 that is not present in the B-homeolog or in ISA1307 or CLIB213T ., In a systematic search , we found a total of 10 A-genes and 9 B-genes that were inactivated only in strain ATCC60483 ( S2 Table ) ., In each case , the other homeolog was intact , and the mutations , discovered in the PacBio assembly , were confirmed by our Illumina contigs of the ATCC60483 genome ., We found a further 8 disabling mutations that are shared between ATCC60483 and ISA1307 ., One of these is the AC-to-AG splice site mutation in the B-homeolog of MATα2 described above ( Fig 5C ) ., Another is the HO endonuclease gene , whose A-homeolog contains an identical 1-bp deletion in both ATCC60483 and ISA1307 , whereas the B-homeolog of HO is intact in both strains ( S2 Table ) ., It is perhaps surprising that the HO gene that degenerated is the A-homeolog , whereas the broken MAT locus is the B-homeolog , but the 2 endonucleases are likely to have had identical site specificities because the HO cleavage site is well conserved among species ., The existence of these 8 shared disabling mutations provides further support for the idea that the 2 strains of Z . parabailii are descended from the same hybrid ancestor , because these mutations may not be viable in the absence of the intact homeologous copies of these genes ., Only one of them is present also in CLIB213T ( S2 Table ) ., We annotated 447 introns in the Z . parabailii ATCC60483 genome , most of which are confirmed by our RNA-Seq data ., There are 428 intron-containing genes , including 19 that have 2 introns ., We did not find any examples of intron presence/absence differences between homeologs ., Interestingly , we found several genes with an in-frame intron—that is , an intron that is a multiple of 3 bp long and contains no stop codons , so that both the spliced and unspliced forms of the mRNA can be translated int

Introduction, Results, Discussion, Materials and methods

Many interspecies hybrids have been discovered in yeasts , but most of these hybrids are asexual and can replicate only mitotically ., Whole-genome duplication has been proposed as a mechanism by which interspecies hybrids can regain fertility , restoring their ability to perform meiosis and sporulate ., Here , we show that this process occurred naturally during the evolution of Zygosaccharomyces parabailii , an interspecies hybrid that was formed by mating between 2 parents that differed by 7% in genome sequence and by many interchromosomal rearrangements ., Surprisingly , Z . parabailii has a full sexual cycle and is genetically haploid ., It goes through mating-type switching and autodiploidization , followed by immediate sporulation ., We identified the key evolutionary event that enabled Z . parabailii to regain fertility , which was breakage of 1 of the 2 homeologous copies of the mating-type ( MAT ) locus in the hybrid , resulting in a chromosomal rearrangement and irreparable damage to 1 MAT locus ., This rearrangement was caused by HO endonuclease , which normally functions in mating-type switching ., With 1 copy of MAT inactivated , the interspecies hybrid now behaves as a haploid ., Our results provide the first demonstration that MAT locus damage is a naturally occurring evolutionary mechanism for whole-genome duplication and restoration of fertility to interspecies hybrids ., The events that occurred in Z . parabailii strongly resemble those postulated to have caused ancient whole-genome duplication in an ancestor of Saccharomyces cerevisiae .

It has recently been proposed that the whole-genome duplication ( WGD ) event that occurred during evolution of an ancestor of the yeast S . cerevisiae was the result of a hybridization between 2 parental yeast species that were significantly divergent in DNA sequence , followed by a doubling of the genome content to restore the hybrid’s ability to make viable spores ., However , the molecular details of how genome doubling could occur in a hybrid were unclear because most known interspecies hybrid yeasts have no sexual cycle ., We show here that Z . parabailii provides an almost exact precedent for the steps proposed to have occurred during the S . cerevisiae WGD ., Two divergent haploid parental species , each with 8 chromosomes , mated to form a hybrid that was initially sterile but regained fertility when 1 copy of its mating-type locus became damaged by the mating-type switching apparatus ., As a result of this damage , the Z . parabailii life cycle now consists of a 16-chromosome haploid phase and a transient 32-chromosome diploid phase ., Each pair of homeologous genes behaves as 2 independent Mendelian loci during meiosis .

fungal spores, chromosome structure and function, centromeres, sequence assembly tools, fungi, model organisms, experimental organism systems, genome analysis, fungal reproduction, saccharomyces, research and analysis methods, genome complexity, mycology, genomics, chromosome biology, genetic loci, yeast, cell biology, genetics, biology and life sciences, yeast and fungal models, saccharomyces cerevisiae, computational biology, introns, organisms, chromosomes

journal.pcbi.1006043

Rational metareasoning and the plasticity of cognitive control

The human brain has the impressive ability to adapt how it processes information and responds to stimuli in the service of high level goals , such as writing an article 1 ., The mechanisms underlying this behavioral flexibility range from seemingly simple processes , such as inhibiting the impulse to browse your Facebook feed , to very complex processes such as orchestrating your thoughts to reach a solid conclusion ., Our capacity for cognitive control enables us to override automatic processes when they are inappropriate for the current situation or misaligned with our current goals ., One of the paradigms used to study cognitive control is the Stroop task , where participants are instructed to name the hue of a color word ( e . g . , respond “green” when seeing the stimulus RED ) while inhibiting their automatic tendency to read the word ( “red” ) 2 ., Similarly , in the Eriksen flanker task , participants are asked to report the identity of a target stimulus surrounded by multiple distractors while overcoming their automatic tendency to respond instead to the distractors ., Individual differences in the capacity for cognitive control are highly predictive of academic achievement , interpersonal success , and many other important life outcomes 3 , 4 ., While exerting cognitive control improves people’s performance in these tasks , it is also effortful and appears to be intrinsically costly 5 , 6 ., The Expected Value of Control ( EVC ) theory maintains that the brain therefore specifies how much control to exert according to a rational cost-benefit analysis , weighing these effort costs against attendant rewards for achieving one’s goals 7 ., In broad accord with the predictions of the EVC theory , previous research has found that control specification is context-sensitive 8 , 9 and modulated by reward across multiple domains 10 , 11 , such as attention , response inhibition , interference control , and task switching ., While previous theories account for that fact that people’s performance in these task is sensitive to reward 7 , 12–14 , it remains unclear how these dependencies arise from people’s experience ., Recently , it has been proposed that the underlying mechanism is associative learning 15 , 16 ., Indeed , a number of studies have demonstrated that cognitive control specification is plastic: whether people exert cognitive control in a given situation , which controlled processes they employ , and how much control they allocate to them is learned from experience ., For instance , it has been demonstrated that participants in visual search tasks gradually learn to allocate their attention to locations whose features predict the appearance of a target 17 , and a recent study found that learning continuously adjusts how much cognitive control people exert in a Stroop task with changing difficulty 18 ., Furthermore , it has been shown that people learn to exert more cognitive control after their performance on a control-demanding task was rewarded 10 and learn to exert more control in response to potentially control-demanding stimuli that are associated with reward than to those that are not 11 ., These studies provide evidence that people can use information from their environment ( e . g . , stimulus features ) to learn when to exert cognitive control and how to exert control , and it has recently been suggested that this can be thought of in terms of associative learning 15 , 16 ., Other studies suggested that cognitive control can be improved through training 19–21 ., However , achieving transfer remains challenging 22–25 , the underlying learning mechanisms are poorly understood , and there is currently no theory that could be used to determine which training regimens will be most effective and which real-life situations the training will transfer to ., Developing precise computational models of the plasticity of cognitive control may be a promising way to address these problems and to enable more effective training programs for remediating executive dysfunctions and enabling people to pursue their goals more effectively ., In this article , we extend the EVC theory to develop a theoretical framework for modeling the function and plasticity of cognitive control specification ., This extension incorporates recent theoretical advances inspired by the rational metareasoning framework developed in the artificial intelligence literature 26 , 27 ., We leverage the resulting framework to derive the Learned Value of Control ( LVOC ) model which can learn to efficiently select control signals based on features of the task environment ., The LVOC model can be used to simulate cognitive control ( e . g . , responding to a goal-relevant target that competes with distractors ) and , more importantly , how it is shaped by learning ., According to the LVOC model , people learn the value of different cognitive control signals ( e . g . , how much to attend one stimulus or another ) ., A key strength of this model is that it is very general and can be applied to phenomena ranging from simple learning effects in the Stroop task to the acquisition of complex strategies for reasoning and problem-solving ., In order to demonstrate the validity and generality of this model , we show that it can capture the empirical findings of five cognitive control experiments on the plasticity of visual attention 17 , the interacting effects of reward and task difficulty on the plasticity of interference control 10 , 11 , and the transfer of such learning to novel stimuli 8 , 9 ., Moreover , the LVOC model outperforms alternate models of such learning processes that rely only on associative learning or a basic win-lose-stay-shift strategy ., Our findings shed light on how learning and experience might shape people’s ability and propensity to adaptively control their minds and behavior , and the LVOC model predicts under which circumstances these mechanisms might lead to self-control failure ., At an abstract level , all cognitive control processes serve the same function: to adapt neural information processing to achieve a goal 28 ., At this abstract level , neural information processing can be characterized by the computations being performed , and the extent to which the brain achieves its goals can be quantified by the expected utility of the resulting actions ., From this perspective , an important function of cognitive control is to select computations so as to maximize the agent’s reward rate ( i . e . , reward per unit time ) ., This problem is formally equivalent to the rational metareasoning 26 , 29 problem studied in computer science: selecting computations so as to make optimal use of the controlled system’s limited computational resources ( i . e . , to achieve the highest possible sum of rewards with a limited amount of computation ) ., Thus , rational metareasoning suggests that the specification of cognitive control is a metacognitive decision problem ., In reinforcement learning 30 , decision problems are typically defined by a set of possible actions , the set of possible states , an initial state , the conditional probabilities of transitioning from one state to another depending on the action taken by the agent , and a reward function ., Together these five components define a Markov decision process ( MDP 30 ) ., In a typical application of this framework the agent is an animal , robot , or computer program , actions are behaviors ( e . g . , pressing a lever ) , the state characterizes the external environment ℰ ( e . g . , the rat’s location in the maze ) , and the rewards are obtained from the environment ( e . g . , pressing a lever dispenses cheese ) ., In general , the agent cannot observe the state of the environment directly; for instance , the rat running through a maze does not have direct access to its location but has to infer this from sensory observations ., The decision problems posed by an environment that is only partially observable can be modelled as a partially observable MDP ( POMDP 31 ) ., For each POMDP there is an equivalent MDP whose state encodes what the agent knows about the environment and is thus fully observable; this is known as the belief-MDP 31 ., Critically , the belief-MDP formalism can also be applied to the choice of internal computations 27–such as allocating attention 32 or gating information into working memory 33 , 34–rather than only physical actions ., In the rational metareasoning framework , the agent is the cognitive control system whose actions are control signals that specify which computations the controlled systems should perform ., The internal state of the controlled systems is only partially observable ., We can formally define the problem of optimal cognitive control specification as maximizing reward the in the meta-level MDP, M= ( S , s0 , C , T , r ) ,, ( 1 ), where S is the set of possible information states , comprising beliefs about the external environment ( e . g . , the choices afforded by the current situation ) and beliefs about the agent’s internal state ( e . g . , the decision system’s estimates of the choices’ utilities ) , s0 denotes the initial information state , C is the set of possible control signals that may be discrete ( e . g . , “Simulate action 1 . ” ) or continuous ( e . g . , “Increase the decision threshold by 0 . 175 . ” or “Suppress the activity of the word-reading pathway by 75% . ” ) , T is a transition model , and r is the reward function that cognitive control seeks to maximize ., The transition model specifies the conditional probability of transitioning from belief state s to belief state s′ if the control signal is c by T ( s , c , s′ ) ., The meta-level reward function r combines the utility of outcome X ( of actions resulting from control signal c in belief state s ) with the computational cost associated with exerting cognitive control:, r ( s , c ) =u ( X ) -cost ( s , c ) ,, ( 2 ), where X is the outcome of the resulting action , u is utility function of the brain’s reward system , and cost ( s , c ) is the cost of implementing the controlled process ., Within this framework , we can define a cognitive control strategy π:S→C as a mapping from belief states s∈S to control signals c∈C ., The optimal cognitive control strategy π⋆ is the one that always chooses the computation with the highest expected value of computation ( EVOC ) :, π⋆:s↦argmaxcEVOC ( c , s ) ., ( 3 ), The EVOC is the expected sum of computational costs and benefits of performing the computation specified by the control signal c and continuing optimally from there on:, EVOC ( c , s ) =Qπ⋆ ( s , c ) =Er ( s , c ) +Vπ⋆ ( St+1 ) |St=s , Ck=c , T ,, ( 4 ), where Qπ⋆ is known as the Q-function of the optimal control strategy π⋆ , and Vπ⋆ ( St+1 ) is the expected sum of meta-level rewards of starting π⋆ in state St+1 ., In summary , cognitive control specification selects the sequence of cognitive control signals that maximizes the expected sum of rewards of the resulting actions minus the cost of the controlled process ., The optimal solution to this problem is given by the optimal control policy π⋆ ., So far , we have assumed that the cognitive control system chooses one control signal at a time , but c could also be a vector comprising multiple control signals ( e . g . , one that increases the rate at which evidence is accumulated towards the correct decision via an attentional mechanism and a second one that adjusts the decision threshold ) ., Furthermore , overriding a habit by a well-reasoned decision also requires executing a coordinated sequence of cognitive operations for planning and reasoning ., Instead of specifying each of these operations by a separate control signal , the cognitive control system might sometimes use a single control signal to instruct the decision system to execute an entire planning strategy ., The rational metareasoning framework allows us to model cognitive strategies as options 35–38 ., An option is a policy combined with an initiation set and a termination condition 38 ., Options can be treated as if they were elementary computations and elementary computations can be interpreted as options that terminate after the first step ., With this extension , the optimal solution to the cognitive control specification problem becomes, π⋆ ( s ) =argmaxo∈OQ⋆ ( s , o ) ,, ( 5 ), where the set of options O may include control strategies and elementary control signals ., Critically , this rational metareasoning perspective on cognitive control covers not only simple phenomena , such as inhibiting a pre-potent automatic response in the Stroop task , but also more complex ones , such as sequencing one’s thoughts so as to follow a good decision strategy , and very complex phenomena such as reasoning about how to best solve a complex problem ., The computations required to determine the expected value of control may themselves be costly and time consuming ., Yet , in some situations cognitive control has to be engaged very rapidly , because maladaptive reflexes , impulses , and habitual responses have to be inhibited before the triggered response has been executed ., In such situations , there is simply not enough time to compute the expected value of control on the fly ., Fortunately , this may not be necessary because an approximation to the EVOC can be learned from experience ., We therefore hypothesize that the cognitive control system learns to predict the context-dependent value of alternative control signals ., By understanding how this learning occurs , we might be able to explain the experience-dependent changes in how people use their capacity for cognitive control , which we will refer to as the plasticity of cognitive control specification ., In addition to these systematic , experience-driven changes cognitive control is also intrinsically variable ., To model the plasticity and the variability of cognitive control , this section develops a model that combines a novel feature-based learning mechanism with a new control specification mechanism that explores promising control signals probabilistically to accelerate learning which of them is most effective ., The previous section characterized the problem of cognitive control specification as a sequential meta-decision problem ., This makes reinforcement learning algorithms 39 a natural starting point for exploring how the cognitive control systems learns the EVOC from experience ., Approximate Q-learning appears particularly suitable because the optimal control strategy can be expressed in terms of the optimal Q-function ( Eqs 3–5 ) ., From this perspective , the plasticity mechanisms of cognitive control specification serve to learn an approximation to the value Qt ( s , c ) of selecting control signal c in state s based on one’s experience with selecting control signals c = ( c1 , ⋯ , ct ) in states s = ( s1 , ⋯ , st ) and receiving the meta-level rewards r = ( r1 , ⋯ , rt ) ., Learning an approximate Q-function Qt from this information could enable the cognitive control system to efficiently select a control strategy by comparing learned values rather than reasoning about their effects ., Learning the optimal meta-level state-value function Q⋆ can be challenging because the value of each control signal may depend on the outcomes of the control signals selected afterwards ., Furthermore , the state space of the meta-level MDP has a very high dimensionality as it comprises all possible states that the controlled system could be in ., To overcome these challenges , a neural system like the brain might learn a linear approximation to the meta-level state value function instead of estimating each of its entries separately ., Concretely , the cognitive control system might learn to predict the value of selecting a control strategy ( e . g . , focusing on the presenting speaker instead of attending to an incoming phone call ) by a weighted sum of features of the internal state and the current context ( e . g . being in a conference room ) ., For instance , the value Q⋆ ( s , c ) of choosing control signal c in the internal state s can be predicted from the features fk ( s ) , the implied control signal intensities c , their interactions with the features , that is fk ( s ) ci , and their costs ., Concretely , the EVOC of selecting control signal c in state s is approximated by the Learned Value of Control ( LVOC ) ,, LVOC ( s , c;w ) =w0+ ( ∑k=1Kwk ( f ) ⋅fk ( s ) ) + ( ∑l=1Lwl ( c ) ⋅cl ) + ( ∑k=1K∑l=1Lwk , l ( f×c ) ⋅fk ( s ) ⋅cl ) −cost ( c ) −w ( T ) ⋅T ,, ( 6 ), where the weight vector w includes the offset w0 , the weights wk ( f ) of the states’ features , the weights w ( c ) of the control signal intensities , the weights wk , l ( f×c ) of their interaction terms , the weight w ( T ) of the response time T , and cost ( c ) is the intrinsic cost of control which scales with the amount of cognitive control applied to the task ., The optimal way to update the weights based on experience in a stationary environment is given by Bayes rule ., Our model therefore maintains and continues to update an approximation to the posterior distribution, P ( w|e1 , ⋯ , t ) ∝P ( w|e1 , ⋯ , t-1 ) ⋅P ( et|w ) ,, ( 7 ), on the weight vector w given its experience e1 , ⋯ , t up until the present time t , where each experience ei = ( si , ci , ri , Ti , si+1 ) comprises the state , the selected control signal , the reward , the response time , and the next state ., In simple settings where a single control signal determines a single reward our model’s learning mechanism is equivalent to Bayesian linear regression 40 , 41 ., In more complex settings involving a series of control signals or delayed rewards the learning rule approximates the Bayesian update by substituting the delayed costs and benefits of control by the model’s predictions ., For more details , see S1 Text ., If the value of control is initially unknown , the optimal way to select control signals is to balance exploiting previous experience to maximize the expected immediate performance with exploring alternative control allocations that might prove even more effective ., Our model solves this dilemma by an exploration strategy similar to Thompson sampling: It draws k samples from the posterior distribution on the weights and averages them , that is, w~1 , ⋯ , w~k∼P ( w|e1 , ⋯ , t ) , w~=1k⋅∑i=1kw~i ., ( 8 ), According to the LVOC model the brain then selects a control signal by maximizing the EVOC predicted by the average weight w~ , that is, ct≈argmaxcLVOC ( st , c;w~ ) ., ( 9 ), Together , Eqs 6–9 define the LVOC model of the plasticity of cognitive control ., The LVOC model extends the EVC theory 7 which defines optimal control signals in terms of the EVOC ( Eq 3 ) , by proposing two mechanisms through which the brain might be able to approximate this normative ideal: learning a feature-based , probabilistic model of the EVOC ( Eqs 6 and 7 ) and selecting control signals by sampling from this model ( Eqs 8 and 9 ) ., This model is very general and can be applied to model cognitive control of many different processes ( e . g . , which location to saccade to vs . how strongly to inhibit the word-reading pathway ) and different components of the same process ( e . g . , rate of evidence accumulation towards the correct decision vs . the decision threshold ) ., The LVOC model’s core assumptions are that the brain learns to predict the EVOC of alternative control specifications from features of the situation and the control signals , and that the brain then probabilistically selects the control specification with the highest predicted value of control ., Both of these components could be implemented by many different mechanisms ., For instance , instead of implementing the proposed approximation to Bayesian regression , the brain might learn to predict the EVOC through the reward-modulated associative plasticity mechanism outlined in the SI ., We are therefore not committed to the specific instantiation we used ( Eqs 7–9 ) for the purpose of the simulations reported below ., The LVOC model instantiates the very general theory that the brain learns how to process information via metacognitive reinforcement learning ., This includes not only the plasticity of cognitive control but also how people might discover cognitive strategies for reasoning and decision-making and how they learn to regulate their mental activities during problem solving ., As a proof of concept , the following sections validate the LVOC model against five experiments on the plasticity of attention and interference control ., In principle , the control-demanding behavior considered in this paper could result from simpler mechanisms than the ones proposed here ., In this section , we consider two simple models that we use as alternatives to compare against the more complex LVOC model ., The first model relies on the assumption that the plasticity of cognitive control can be understood in terms of associative learning 15 , 16 ., We therefore evaluate our model against an associative learning model based on the Rescorla-Wagner learning rule 42 ., This model forms stimulus-control associations based on the resulting reward ., The association As , c between a stimulus s and a control signal c is strengthened when it is accompanied by ( intrinsic or extrinsic ) reward and weakened otherwise ., Concretely , the association strengths involving the chosen response were updated according to the Rescorla-Wagner rule , that is, As , c=As , c+α⋅ ( R-∑sIs⋅As , c ) ,, ( 10 ), where α is the learning rate , R is the reward and the indicator variable Is is 1 when the stimulus s was present and 0 else ., Given the learned associations , the control signal is chosen probabilistically according to the exponentiated Luce’s choice rule , that is each control signal c is selected with probability, p ( c ) =exp ( As , c ) ∑cexp ( As , c ) ., ( 11 ), The second alternative model is based on previous research suggesting that people sequentially adjust their strategy through a simple Win-Stay Lose-Shift mechanism 43 ., On the first trial , this mechanism chooses a strategy at random , and on each subsequent trial it either repeats the previous strategy when it was successful or switches to a different strategy when the current strategy failed ., Here , we apply this idea to model how the brain learns which control signal to select ., Concretely , our WSLS model repeats the previous control signal ( e . g . , “Attend to green . ” ) when it leads to a positive outcome ( Win-Stay ) and randomly selects a different control signal ( e . g . , “Attend to red . ” ) otherwise ( Lose-Shift ) ., In contrast to the LVOC mode , the two alternative models assume that control signals are discrete rather than continuous ., In the context of visual attention , they choose their control signal c from the set {1 , 2 , 3 , ⋯ , 12} of possible locations to attend , and in the context of inhibitory control they decide to either inhibit the process completely or not at all ( c ∈ {0 , 1} ) ., To evaluate the proposed models , we used them to simulate the plasticity of attentional control in a visual search task 17 as well as learning and transfer effects in Stroop and Flanker paradigms 8–11 ., Table 1 summarizes the simulated phenomena and how the LVOC model explains each at a conceptual level ., Lin et al . 17 had participants perform a visual search task for which the target of attention could either be predicted ( training and predictable test trials ) or not ( unpredictable test trials ) ( Fig 1a ) ., For this task , given its core reinforcement learning assumption ( Table 1 ) , the LVOC model predicts that, 1 ) people should learn to attend to the circle with the predictive color and thus become faster at finding the target over the course of training ,, 2 ) continue to use the learned attentional control strategy in the test block and hence be significantly slower when the target appears in a circle of a different color during the test block , and, 3 ) gradually unlearn their attentional bias during the test block ( Fig 1c ) ., As shown Fig 1b , all three predictions were confirmed by Lin and colleagues 17 ., We compared the performance of LVOC to two plausible alternative models of these control adjustments: a Win-Stay Lose-Shift model and a simple associative learning model based on the Rescorla-Wagner learning rule ., We found that the Win-Stay Lose-Shift model failed to capture that people’s performance improved gradually during training , and it also failed to capture the difference between people’s response times to predicted versus unpredicted target locations in the test block ( see Fig 1d ) ., As Fig 1e shows , the fit of the associative learning model ( estimated learning rate: 0 . 0927 ) captures that after learning to exploit the predictive regularity in the training block participants were significantly slower in the test block ., However , this simple model predicted significantly less learning induced improvement and significantly slower reaction times than was evident from the data by 17 ., A quantitative model comparisons using the Bayesian Information Criterion 60 , 61 provided very strong evidence that the LVOC model explains the data by 17 better than the Rescorla-Wagner model or the Win-Stay Lose-Shift model ( BICLVOC = 1817 . 8 , BICRW = 9763 . 2 , BICWSLS = 3449 . 9 ) ., This reflects that our model was able to accurately predict the data from 17 without any free parameters being fitted to those data ., In conclusion , findings suggest that the LVOC model correctly predicted essential learning effects observed by 17 and explains these data significantly better than a simple associative learning model and a Win-Stay Lose-Shift model ., To more accurately capture both the slow improvement in the training block and the rapid unlearning in the test block simultaneously , the LVOC model could be extended by including a mechanism that discounts what has been learned or increases the learning rate when a change is detected 62 , 63 ., Next , we evaluate the LVOC model against empirical data on the plasticity of inhibitory control ., We found that our model can capture reward-driven learning effects in Stroop and Flanker tasks , as well as how people learn to adjust their control allocation based on features that predict incongruence and the transfer of these learning effects to novel stimuli ., In each case , the LVOC model captured the empirical phenomenon more accurately than either a simple Win-Stay Lose-Shift model or a simple associative learning model ., The following two sections present these results in turn ., The expected value of computation depends not only on the rewards for correct performance but also on the difficulty of the task ., In easy situations , such as the congruent trials of the Stroop task , the automatic response can be as accurate , faster , and less costly than the controlled response ., In cases like this , the expected value of exerting control is less than the EVOC of exerting no control ., By contrast , in more challenging situations , such as incongruent Stroop trials , the controlled process is more accurate and therefore has a positive EVOC as long as accurate performance is sufficiently important ., Therefore , on incongruent trials the expected value of control is larger than the EVOC of exerting no control ., Our model thus learns to exert control on incongruent trials but not on congruent trials ., Our model achieves this by learning to predict the EVOC from features of the stimuli ., This predicts that people should learn to exert more control when they encounter a stimulus feature ( such as a color or word ) that is predictive of incongruence than when they encounter a feature that is predictive of congruence ( see Table 1 ) ., Consistent with our model’s predictions , Bugg and colleagues 8 found that people learn to exert more control in response to stimulus features that predict incongruence than stimulus features that predict congruence ., Their participants performed a color-word Stroop task with four colors and their names printed either in cursive or regular font ., Our model captured the effects of congruency-predictive features on control allocation with a plausible set of parameters ( see Table 2 ) ., As shown in Fig 4a and 4b , the LVOC model predicted that responses should be faster ( 655 ± 9 ms vs . 722 ± 11 ms; t ( 49 ) = 5 . 39 , p < 0 . 0001 ) and more accurate ( 2 . 85 ± 0 . 2% errors vs . 4 . 3 ± 0 . 3% errors; t ( 49 ) = 5 . 01 , p < 0 . 0001 ) on incongruent trials if the word was predictive of incongruence than when it was not ., To their surprise , Bugg and colleagues observed that adding an additional feature ( font ) that conveyed the same information about congruence as the color , did not enhance learning ., This is exactly what our model predicted because the presence of a second predictive feature reduces the evidence for the predictive power of the first one and vice versa–this is directly analogous to a phenomenon from the Pavlovian literature known as blocking , whereby an animal fails to learn an association between a stimulus and an outcome that is already perfectly predicted by a second stimulus 64 ., Since our model learns about the predictive relationship between features and the EVOC , it predicts that all learning effects should transfer to novel stimuli that share the features that were predictive of the expected value of control in the training trials ( see Table 1 ) ., A separate study by Bugg and colleagues 9 confirmed this prediction ., They trained participants in a picture-word Stroop task to associate particular images of certain categories ( e . g . , cats and dogs ) with incongruence and associated particular images of other categories ( e . g . , fish and birds ) with congruence ., As expected , participants learned to exert more control when viewing the stimuli associated with incongruence ., More importantly , these participants also exerted more control when tested on novel instances of the category associated with incongruence ( e . g . , cats ) than on novel instances of the category associated with congruence ( e . g . , fish ) ., This finding provides strong evidence for the feature-based learning mechanism that is at the core of our model of the plasticity of cognitive control and is entirely accounted for by our model ., As shown in Fig 4e and 4f , our model correctly predicted the positive and the negative transfer effects reported by 9 with reasonable parameters ( see Table 2 ) : The model’s responses were faster ( 709 ± 3 ms vs . 685 ± 2 ms; t ( 99 ) = −8 . 13 , p < 0 . 0001 ) and more accurate ( 4 . 8 ± 0 . 3% errors vs . 3 . 2 ± 0 . 1% errors; t ( 99 ) = −5 . 06 , p < 0 . 0001 ) on incongruent trials if the word was predictive of incongruence than when it was not ( positive transfer ) ., Conversely , on congruent trials , the predicted responses were slightly slower when the features wrongly predicted incongruence ( 527 ± 0 . 2ms vs . 530 ± 0 . 1ms , t ( 99 ) = 9 . 28 , p < 0 . 0001; negative transfer ) ., The model developed in this article builds on two previous theories: the EVC theory , which offered a normative account of control specification 7 , and the rational metareasoning theory of strategy selection 53 , which suggested that people acquire the capacity to select heuristics adaptively by learning a predictive model of the execution time and accuracy of those heuristics ., The LVOC model synergistically integrates these two theories: it augments the EVC theory with the metacognitive learning and prediction mechanisms identified by 53 , and it augments rational metareasoning models of strategy selection with the capacity to specify continuous control signals that gradually adjust parameters of the controlled process ( see S2 Text ) ., All else being equal , the proposed learning rules ( see Eq 7 , S1 Text Equations 1–7 , and S3 Text Equations 13–14 ) predict that people’s propensity to exert cognitive control should increase when the controlled process was less costly ( e . g . , faster ) or generated more reward than expected

Introduction, Models, Results, Discussion

The human brain has the impressive capacity to adapt how it processes information to high-level goals ., While it is known that these cognitive control skills are malleable and can be improved through training , the underlying plasticity mechanisms are not well understood ., Here , we develop and evaluate a model of how people learn when to exert cognitive control , which controlled process to use , and how much effort to exert ., We derive this model from a general theory according to which the function of cognitive control is to select and configure neural pathways so as to make optimal use of finite time and limited computational resources ., The central idea of our Learned Value of Control model is that people use reinforcement learning to predict the value of candidate control signals of different types and intensities based on stimulus features ., This model correctly predicts the learning and transfer effects underlying the adaptive control-demanding behavior observed in an experiment on visual attention and four experiments on interference control in Stroop and Flanker paradigms ., Moreover , our model explained these findings significantly better than an associative learning model and a Win-Stay Lose-Shift model ., Our findings elucidate how learning and experience might shape people’s ability and propensity to adaptively control their minds and behavior ., We conclude by predicting under which circumstances these learning mechanisms might lead to self-control failure .

The human brain has the impressive ability to adapt how it processes information to high level goals ., While it is known that these cognitive control skills are malleable and can be improved through training , the underlying plasticity mechanisms are not well understood ., Here , we derive a computational model of how people learn when to exert cognitive control , which controlled process to use , and how much effort to exert from a formal theory of the function of cognitive control ., Across five experiments , we find that our model correctly predicts that people learn to adaptively regulate their attention and decision-making and how these learning effects transfer to novel situations ., Our findings elucidate how learning and experience might shape people’s ability and propensity to adaptively control their minds and behavior ., We conclude by predicting under which circumstances these learning mechanisms might lead to self-control failure .

learning, control theory, engineering and technology, social sciences, neuroscience, learning and memory, control engineering, cognitive psychology, systems science, mathematics, cognition, vision, computer and information sciences, human learning, psychology, control systems, biology and life sciences, physical sciences, sensory perception, cognitive science, attention

journal.pgen.1002820

Large-Scale Introgression Shapes the Evolution of the Mating-Type Chromosomes of the Filamentous Ascomycete Neurospora tetrasperma

Introgression is a process by which two species mate and produce a hybrid offspring , after which the hybrid offspring repeatedly backcross with one of the parental species ., It ultimately results in genetic material from one species infiltrating another , genetically differentiated , species 1 ., Introgression is a key process in evolution , as it may contribute to speciation , diversification and adaptation to new environments 1 , 2 ., The importance and prevalence of introgression has been well established in plant systems ., For example , Rieseberg ( 1993 ) summarized 65 cases of introgression associated with plant speciation 2 ., In addition , Mallet ( 2005 ) estimated that up to 25% of plant species produce viable offspring from interspecific matings , which lead to simple hybridization and introgression 3 ., In the animal kingdom , introgression has also been recognized as an important factor driving genome evolution 3–7 , possibly transferring key genes for development among species 8 , 9 ., Evidence suggests that introgression does not occur randomly across genomes ., For instance , introgression may be less likely to occur in genomic regions with complex hybrid incompatibility loci 1 , such as sex chromosomes in heterogametic organisms ( e . g . , XY in mammals 10 ) ., This phenomenon is consistent with Haldanes rule and the large X effect hypothesis , which predicts that the density of hybrid incompatibility loci will be greatest on the X chromosome in animals 11 ., Introgression is also less likely to occur in regions of suppressed recombination 1 , 12 , 13 ., Nonetheless , simulation studies have shown that if the hybrid fitness is not exceedingly low , even slightly deleterious introgressions have a good probability of fixation , due to genetic drift or population demographics 14–16 ( see also 17–20 ) , and thus might not always be highly dependent on genomic location ., In the fungal kingdom , the role of introgression is beginning to be explored ., In the past five years , interspecific gene flow , especially of virulence genes , has been found in multiple fungal systems , e . g . , Coniophora , Fusarium , Microbotryum , Aspergillus , Heterobasidion and Stagnospora 21–27 ., In these cases , introgression has provided adaptive advantages to pathogenic species ., A notable pattern of introgression is also found in Neurospora , Ophiostoma and Stemphylium 28–30 ., Reproductive genes in these systems seem to be more permeable to introgression than housekeeping genes or non-coding loci , which is in contrast to theoretical expectations 11 ., Most studies of interspecific gene flow in fungi are based on multilocus data , thus being unable to reveal the actual size , abundance and distribution of introgression tracts on the genomic level ., So far , only a limited number of studies have used genomic approaches to study introgression , and these studies revealed small , discrete introgression tracts in Coccidioides posadasii ( 70 introgressed regions of which 60 regions <50 kb ) , Aspergillus fumigatus ( 189 introgressed regions , 1 Mbp total size ) and N . crassa ( 2 introgressed regions , each ∼10 kb ) 31–33 ., The growing number of genomic datasets from the fungal kingdom provides the potential to increase our understanding of introgression in fungal genomes 34 ., Furthermore , the fact that fungi have no differentiated sexes , i . e . , female/male dichotomy of individuals carrying gametes of different sizes , make them alternative , and simple , models to study general processes in nature expected to be affected by sex-biased evolutionary forces 35 ., The filamentous ascomycete genus Neurospora is a particularly useful model system to study introgression in nature ., The heterothallic ( self-incompatible ) and pseudohomothallic ( partially self-compatible ) species of Neurospora constitute the terminal clade in the genus phylogeny 36 ., This clade contains several well-characterized , closely related species , which are all able to sexually outcross in nature 36 ., The heterothallic species , represented by the model species N . crassa , grow vegetatively as haploid and require a partner with nuclei of a compatible mating type ( mat A or mat a ) in order to go through the sexual cycle ., In contrast , the pseudohomothallic N . tetrasperma grows as heterokaryotic for mating type ( i . e . , cells harbor haploid nuclei of opposite mating types ) and are thereby predominantly self-fertile ., Neurospora tetrasperma also rarely forms homokaryotic , single mating-type individuals that undergo outcrossing events to return to the heterokaryotic stage 37 , 38 ., Although the species in the terminal Neurospora clade are reproductively isolated , overlapping geographical distributions may have created opportunities for hybridization between them 36 , 39 , 40 , and indeed , fertile interspecies crosses in the laboratory have been reported 38 , ., Phylogenetic studies by Skupski et al . ( 1997 ) and Strandberg et al . ( 2010 ) 28 , have shown different relationships between Neurospora species for genes on the autosomes and the mat genes , possibly reflecting introgression between species ., The pseudohomothallic N . tetrasperma has attracted considerable research attention ., The genome of this taxon is highly syntenic to the model species N . crassa , which harbors 7 chromosomes in 41 Mbp ( http://www . broadinstitute . org ) ., However its mating-type ( mat ) chromosomes contain a large region of suppressed recombination ( 75% of the chromosome , or >5 Mbp ) 35 , 37 , 43 , 44 ., Consequently , N . tetrasperma provides a model system to study the evolution of sex chromosomes 35 ., The suppressed recombination region evolved 3 . 5–5 . 8 MYA ago 35; this is in a similar range to that of other young sex chromosomes ( e . g . , Silene 45–48 ) ., A recent analysis of high quality genomic data from two N . tetrasperma homokaryotic strains ( FGSC 2508 and FGSC 2509 ) of opposite mating type revealed that the mat A chromosome has experienced a history of three inversions that encompasses the majority of the region of suppressed recombination , while the mat a chromosome is collinear with the N . crassa mat chromosome 49 ., Recent molecular evolution studies have revealed genetic degeneration in the region of suppressed recombination in N . tetrasperma , as evidenced by reduced codon usage bias and the accumulation of non-synonymous substitutions , as compared to the flanking , recombining , regions of the chromosomes 49–51 ., These data are consistent with genomic degeneration in the young regions of suppressed recombination , similar to trends reported for ancient eukaryotic sex chromosomes 45 , 48 , 52–54 ., Notably , asymmetrical degeneration between the two mat chromosomes has been found 49 , 51 , which was suggested by Whittle et al . ( 2011 ) 51 to be caused by factors such as chromosome-specific structural rearrangements on the mat A chromosome 44 , 49 and/or rare outcrossing or interspecific hybridization events 38 , 55 ., Neurospora tetrasperma was first described by morphological characters 56 , and several studies have indicated that the pseudohomothallic mating system of N . tetrasperma is derived from true heterothallism , and that it is monophyletic 38 , 42 , 57–59 ., Furthermore , N . tetrasperma has recently been recognized as a species complex consisting of multiple genetically and largely reproductively isolated lineages , for which the relationship is largely unresolved 38 ., The aim of the present study was to use comparative genomics of Neurospora species to investigate the introgression landscape at the genomic level ., We acquired medium coverage genomic data from multiple N . tetrasperma lineages , and by using interspecific genomic comparisons with N . crassa we revealed the abundance , size and distribution of introgression tracts among the species ., We used multilocus genealogies from additional heterothallic species of Neurospora to infer the direction of introgression events ., Finally , we investigated patterns of molecular evolution in the introgressed regions ., We sequenced the mat A and mat a haploid genomes from three wild-type heterokaryotic strains of N . tetrasperma , selected to represent three genetically and reproductively isolated lineages 38 ., The six haploid genomes are referred to herein by their lineage ID followed by mating type , i . e . , L1A , L1a , L4A , L4a , L9A and L9a ( Table S1 ) ., Approximately 7 million illumina paired-end reads were generated and mapped to the reference genome of N . crassa ( release version 10 , ∼41 MB ) , yielding an ∼15-fold medium coverage data for each of the six haploid genomes ( Table S2 ) ., Reads covered ∼80% of the genomes , and were evenly distributed ( coverage depth 10–20X ) , except repeat-enriched centromeric regions ( coverage depth <5X ) ( Table S2 , Figure S1 ) ., The seven assembled chromosomes are referred to as linkage group I ( LGI , also referred to as the mat chromosome ) , LGII , LGIII , LGIV , LGV , LGVI , LGVII , as in N . crassa 60 ., We inferred the phylogenetic relationship of our selected N . tetrasperma lineages and N . crassa , using a concatenated dataset of 1 , 978 autosomal genes with the maximum likelihood method ., The tree topology ( Figure 1 ) shows that lineage 4 ( RLM131 ) represents a slightly earlier diverging lineage than lineage 1 ( P4492 ) and lineage 9 ( 965 ) ., Importantly , for the purpose of this discussion , the autosomal alleles of the two strains of different mating types from each lineage ( e . g . , L9A and L9a ) were in all cases nearly identical ., We investigated the sequence divergence between pairs of chromosomes of the same lineage of N . tetrasperma , and found strikingly different patterns for the autosomes and the mat chromosome ., The overall chromosomal divergence levels between the haploid genomes ranged from 3 . 7×10−3 to 5 . 5×10−3 substitutions per site for the six autosomes ( Table S3 ) , which is 2–4 fold lower than the overall divergence between the mat chromosomes , for which we found 1 . 0×10−2 to 1 . 6×10−2 substitutions per site ( Table S3 ) ., To study the variation in sequence divergence across the chromosomes for haploid genomes within lineages , we used a sliding window approach , with a window size of 500 kb and a step size of 100 kb ., For the autosomes , divergence levels were relatively homogenous across the chromosomes , with the exception of slight increases at the flanking left ends of LGII , III , IV and V , at the flanking right end for autosome VI and in the central segment of autosome VII ( Figure S2A ) ., However , even at the peak , the overall levels of divergence did not exceed 1% in any of the autosomes ., The regions of increased divergence also showed lower G+C content levels; 44–48% , as compared to more than 50% in all other regions of the autosomes ( Figure S2C ) ., Previous studies of LGI , LGIII and LGVII in N . crassa 61 , 62 suggest that these sites correspond to the centromeres ., The sequence divergence profile of the mat chromosomes was markedly different from those of the autosomes in that the central portion showed 5 to 10-fold higher divergence levels ( Figure 2A ) ., As in the autosomes , we recorded a short region of lower G+C level for the mat chromosomes ( Figure 2C ) , presumably at the site of the centromere ., However , the segment showing increased divergence levels was much larger , suggesting that this pattern is not explained by the location of centromeres ., On the right and left flanks of the mat chromosomes , the intra-lineage divergence levels dropped to less than 1% ( Figure 2A ) , which is a similar level as observed for the autosomes ( Figure 2 , Figure S2A ) ; these flanking regions of the mat chromosomes are referred to as pseudoautosomal ( PA ) regions ., The size of the region with divergence levels above 1% was estimated to be at least 5 . 37 , 5 . 86 and 5 . 88 Mb in lineages L1 , L4 and L9 , respectively ( Table S4 ) , covering about 70–77% of the chromosome size ., The overall divergence level of the central part of the mat chromosome differed between each lineage ( Figure 2A , Table S4 , L1: 0 . 0187 , L4: 0 . 0238 , L9: 0 . 0343 ) , as did the border positions to the pseudoautosomal ( PA ) regions ( Figure 2A , Table S4 ) , suggesting that the mat chromosomes have independent evolutionary histories in the three lineages ., Atypically high nucleotide sequence divergence levels between the mat chromosomes within a lineage could be due to transfer of DNA from another species ( i . e . , introgression ) or to reduced rates of sequence exchange across the two mat chromosomes ( i . e . , suppressed recombination ) ., In the first case , we expect one of the two chromosomes to show a closer relationship to a mat chromosome of another species , e . g . N . crassa , than to its homolog of the same lineage ., However , under the second hypothesis , we expect the two homologous chromosomes of each lineage to show the same divergence to other species ., Furthermore , reduced sequence exchange by suppressed recombination is expected to lead to the same divergence between the mat A and the mat a chromosomes of N . tetrasperma , i . e . , the mat A chromosomes should be as similar to each other as the mat a chromosomes ., To test these hypotheses , we examined the level of sequence similarity of each N . tetrasperma genome to all other N . tetrasperma genomes included herein , and to the outgroup species N . crassa ., For the autosomes , as expected , the nucleotide divergences of the N . tetrasperma haploid genomes of the same lineage ( Figure S2A , Table S3; 0 . 0045±0 . 0006 ) were considerably lower than divergences obtained by inter-lineage comparisons among genomes of N . tetrasperma ( dotted and dashed lines of Figure S2B , Table S3; 0 . 0208±0 . 0016 ) , which , in turn , were lower than inter-species divergences of N . tetrasperma and N . crassa genomes ( solid lines of Figure S2B , Table S3; 0 . 0423±0 . 00054 ) ( Table S3 ) ., In support of introgression , the pattern of divergence obtained from the central region of the mat a chromosomes stands in striking contrast to that of the autosomes ., In L9 , comparison of sequence divergence in the central region of the mat chromosome showed a lower level of divergence when the L9 mat a strain of N . tetrasperma ( i . e . , L9a ) was compared to N . crassa than when L9a was compared to L9A ( 0 . 027 vs 0 . 034; Figure 2 , Tables S4 , S5 ) ., This difference in divergence was supported by a shared-nucleotide test 6 , which revealed a significantly higher number of uniquely shared nucleotides between L9a and N . crassa than between L9A and N . crassa in this region ( 122 , 510 vs . 74 , 478 , Binomial Sign Test P<10−10; nt identity , L9a-N . crassa 97 . 3% , L9A-N . crassa 96 . 4% ) ., This pattern was not found for any other of the N . tetrasperma sequences ( L1 , L4 ) in comparison with N . crassa ., However , in support for introgression of all lineages , the pattern of divergence of the mat A chromosomes stands in contrast to that of the mat a chromosomes ., Specifically , the divergence levels of all three pair-wise comparisons among the mat A chromosomes were lower than the divergences between the mat a chromosomes of L1 , L4 and L9 ( L1A–L9A: 0 . 0169 , L1A–L4A: 0 . 0166 , L4A–L9A: 0 . 0183 , L1a–L9a: 0 . 0360 , L1a–L4a: 0 . 0240 , L4a–L9a: 0 . 0370 ) ., This difference in divergence was supported by a shared-nucleotide test , which revealed that the mat A-mat A chromosomes shared more unique nucleotides than mat a-mat a in all pair-wise comparisons ( for example , L9 mat A\u200a=\u200aL4 mat A≠L4 mat a 82 , 774 vs . L9 mat a\u200a=\u200aL4 mat a≠L4 mat A 61 , 973 , P<1×10−10 , Binomial Sign Test ) ., Taken together , these results indicate infiltration of foreign DNA , i . e . , introgression , into the mat a chromosomes of all investigated lineages of N . tetrasperma ., We determined the size of the introgression tracts of the mat a chromosomes as follows ., In L9 we estimated the tract to be 5 . 6 Mbp ( 73% of the mat a chromosome ) by nucleotide difference comparison of L9A-N . crassa and L9a-N ., crassa , in L1 to 4 . 1 Mbp ( 53% of the mat a chromosome ) by comparing L1A–L9A and L1a–L9A , and in L4 to 5 . 2 Mbp ( 68% of the mat a chromosome ) by comparing L4A–L9A and L4a–L9A ., We chose N . crassa or L9A as reference strains based on the divergence patterns of Figure 2B ., The estimated borders of the introgression tracts are shown in Table S4 , and schematically depicted in Figure 3 ., For each lineage , the introgression tract was confined to the region of elevated divergence ( Figure 3 , Table S4 ) ., In the mat a chromosomes of L9 and L4 the introgression tracts extended the majority of the divergent region , while in L1 , the introgression tract is 1 . 28 Mbp shorter than the region of elevated divergence ( Figure 3 , Table S4 ) ., The right border of the introgression tract is shared between L9 and L4 , while the left border differs slightly for all lineages ( Figure 3 , Table S4 ) ., We refer to the region that is introgressed in the mat a chromosomes of all three heterokaryotic lineages as Region I and the right flank region introgressed in L4 and L9 , but not in L1 , as Region II ., The region on the left flank , which shows a variable pattern of introgression and divergence among the three lineages , is referred to as Region III ( Figure 3 ) ., To identify the origin and direction of introgression , we carried out phylogenetic inferences of lineage relationships for single loci located on the autosomes and the mat chromosome ( 10 autosomal loci and 16 loci on the mat chromosomes , Figure 3 , Table S6 ) ., For four microsatellite flanking loci located on autosomes ( Table S6 ) and four gene loci evenly distributed across the mat chromosomes ( highlighted in bold in Figure 3 ) , we included in the analyses all currently identified heterothallic species and subgroups of Neurospora ( Table S1 , trees shown in Figure 4 ) , while for the additional loci a subset of heterothallic species were included ( Table S6 , trees shown in Figure S3 ) ., The results revealed that for all autosomal genes , the two haploid single mating-type components of each lineage clustered together ( Figure 4 , Figure S3 ) , as also observed in the tree topology from concatenated genes of the autosomes ( Figure 1 ) ., Likewise , the genes located on the PA flanking ends of the mat chromosomes show near identity in comparisons of alleles from mat A and mat a haploid genomes ( trees of nit-1 and phr are shown in Figure 5 , and ro-10 , mus-42 , prd-4 in Figure S4 ) , consistent with homogenization by recombination ., In contrast , phylogenies inferred from genes located in the central region of the mat chromosomes showed no clustering of the mat A and mat a chromosomes from the same lineage ( Figure 5 ) ., For both of eth-1 and cys-9 , L9a clustered with N . crassa subgroup C and N . perkinsii ( bootstrap support 84-97% ) , and the sequence similarity between L9a and N . crassa ( NcC , strain 8863 ) and N . perkinsii ( strain 8835 ) is 99 . 2% and 99 . 6% , respectively , for these two genes ., L4a clustered with N . hispaniola ( 100% bootstrap support ) , and showed a very high sequence similarity with strain 8817 ( 99 . 8% ) ., These patterns were supported by the additional trees of genes from the mat chromosome , for which only a subset of the heterothallic taxa were included ( Figure S4 ) ., Note that in L4 , for genes between rid-1 and upr-1 ( i . e . , region III , Figure S4 ) the alleles from the mat A chromosome ( and not the mat a chromosome ) clustered together with N . hispaniola ., A possible explanation for this observed discord is the occurrence of a relatively recent crossover event between the mat chromosomes in L4 , at a chromosomal location between upr-1 and arg- 1 ., Indeed , evidence for a crossover on the mat chromosome at this location for L4 has previously been shown 63 ., Furthermore , for the genes of region III , in which the divergence indicate no introgression of L4a ( Figure 3 ) , gene tree analyses support introgression from N . hispaniola , indicating that the divergence method is likely to underestimate the size of the introgression tract ., Finally , in the gene lys-4 , both alleles of L4 cluster together , consistent with a gene conversion event reported for this gene in this lineage 63 ., To quantify the amount of gene tree discordance for the different genomic regions , we performed a Bayesian concordance analysis ( BCA ) of the 26 loci of Table S6 , using the program BUCKy v1 . 4 . 0 64 ., A BCA analysis infers concordance factors as a measure of the proportion of loci that support a given bipartition of the data 65 ., Here we employed the BCA to see if we observe a different inferred history from autosomal loci versus mat chromosome loci , as such a non-random pattern may be due to the action of introgression rather than incomplete lineage sorting ., The analysis was performed using all included lineages of N . tetrasperma , and representatives of the four heterothallic species N . crassa , N . sitophila , N . hispaniola and N . discreta ( Table S1 , Table S6 ) ., In Table S7 and Table S8 we show the sample-wide concordance factors ( CF ) and 95% credible intervals for clades in the analysis of autosomal and mat chromosome loci , respectively , and in Figure 6 we present a plot of the sample-wide CFs for three conflicting clades of specific interest: the L9-N ., crassa clade ( red ) , the L4-N ., hispaniola clade ( blue ) and the N . tetrasperma clade ( a monophyletic grouping of all six N . tetrasperma lineages; green ) in different genomic regions ., We found that the estimate of CF is robust to the change of the prior alpha ( a test of the effect of changing the prior is shown in Figure S5 ) and the results presented here are from the analysis with the prior set to 1 ., For the autosomal loci , the dominant history observed is the N . tetrasperma clade with a CF of 0 . 291 ( 95% credible interval 0 . 1–0 . 5 ) , while the L9-N ., crassa clade and the L4-N ., hispaniola clade received little or no support ( CF<0 . 0001 for both clades: Figure 6 , Table S7 , Figure S6 ) ., Other clades that conflicted with the monophyly of N . tetrasperma ( Table S7 ) showed lower CF values; however , their 95% credibility intervals did overlap with the N . tetrasperma clade ( e . g . , 1 , 2 , 5 , 7 , 8|3 , 4 , 6 , 9 , 10: Table S7 ) ., This overlap means that for the autosomes we cannot statistically designate the N . tetrasperma monophyly as being the dominant history for the N . tetrasperma lineages , although the CF is highest for this clade when compared to those clades that are in conflict with it ., Consistent with introgression of the mat a chromosome of the N . tetrasperma L9 from N . crassa , we observed a high CF on the mat a chromosome for the L9a-N ., crassa clade ( CF 0 . 616 , 95% credible interval 0 . 562–0 . 688 ) , while no support for this clade on the mat A chromosome or autosomes was found ( Figure 6 ) ., This pattern is less apparent for the L4-N ., hispaniola clade ., The CF for this clade is relatively high for both mat chromosomes , i . e . , it is noticeably higher than expected for the mat A chromosome ( 0 . 453 ( 0 . 375–0 . 5 ) ) : Figure 6 , Table S8 ) ., A possible explanation for this observed discord is the crossover event between the mat chromosomes in L4 previously reported by Menkis et al . ( 2010 ) 63 ., To further investigate the transfer of genetic material between mat chromosomes of L4 as the reason for the discordance , we partitioned our data into loci that are to the left and right of the location of the crossover event ( proposed by Menkis et al . ( 2010 ) to be between markers upr-1 and arg-1 ( Figure 3 ) 63 ) , and carried out BCA analyses of the two separate datasets ., The plots of the CFs to the opposite sides of the inferred crossover event support the view that a crossover event has transferred a part of the introgressed region from the mat a to the mat A chromosome; the CF for the RLM131-N ., hispaniola clade is higher on the left of the crossover on the mat A and lower to the right , while the opposite pattern is observed for the mat a chromosome ( Figure 7 ) ., Taken together , the results show that the pattern of phylogenetic discordance on the mat a chromosome is distinct from the patterns observed for the mat A chromosome and the autosomes , and provides further support for large-scale introgression tracts on the mat a chromosome of the N . tetrasperma lineages under study ., In order to reveal whether the previously reported signs of asymmetrical degeneration between the two mat chromosomes in N . tetrasperma 49 , 51 can be explained by introgression of large chromosomal regions from freely recombining heterothallic species , we analyzed substitution frequencies at nonsynonymous and synonymous sites , and codon usage patterns , in a concatenated data set of complete coding sequences ( CDS ) for 543 genes from the region of the mat chromosomes subjected to introgression in all investigated lineages of N . tetrasperma ( Region I , Figure 3 ) ., We expected the degeneration to be reduced in introgressed regions , and followed the method developed previously for studying molecular degeneration in N . tetrasperma , for which an increase in ω ( dN/dS ) was verified to be due to the accumulation of slightly deleterious mutations and that the accumulation of non-preferred codons were not due to a mutational bias 50 , 51 ., Consistent with the expectations , an elevated ω on mat A chromosomes , as compared to the mat a chromosomes , was revealed by using branch models of ω implemented in the codeml program of the Phylogenetic Analysis by Maximum Likelihood ( PAML ) package 66 on the concatenated gene sequence data ., First , we estimated ω for each branch in the phylogeny by running the free ratio model ., The result from this test indicated that the three branches delineating the mat A chromosomes have higher ω ( L9A: ω\u200a=\u200a0 . 2005; L1A: ω\u200a=\u200a0 . 1937; L4A: ω\u200a=\u200a0 . 1976 ) than the mat a branches ( L9a: ω\u200a=\u200a0 . 1450; L1a: ω\u200a=\u200a0 . 1696; L4a: ω\u200a=\u200a0 . 1562 ) ., The ω value for the branch of the freely recombining heterothallic species , N . crassa , was 0 . 1229 , which was the lowest among all seven branches in the input phylogeny ., Furthermore , within each lineage , we found a significantly better fit for a local model of ω allowing a different ω on mat A chromosomes than mat a chromosomes ( P<1×10−10 for L9 , P\u200a=\u200a3 . 7×10−3 for L1 , P\u200a=\u200a1 . 5×10−8 for L4: in each of these cases ω was higher in the mat A chromosomes ) ., Finally , a local branch model in which the three branches delineating mat A chromosomes and the three mat a branches were allowed to have a separate ω , fitted the data significantly better than the global model ( P<1×10−10 , mat A branches ω\u200a=\u200a0 . 1853 , mat a branches ω\u200a=\u200a0 . 1474 ) ., We assessed the accumulation of non-preferred ( NPR ) codons relative to preferred ( PR ) codons in the 543 genes for all six haploid genomes , to test if asymmetrical degeneration can be detected on the level of synonymous codon substitution ., We found a net accumulation of NPR relative to PR in all six N . tetrasperma haploid genomes ( Binomial Sign test , p<1×10−10 ) ( Table 1 ) ., Also , we report statistically significantly higher number of NPR substitutions ( than PR ) in L1A and L4A as compared to L1a and L4a ( Binomial Sign test , P\u200a=\u200a1 . 6×10−10 for L4 , P\u200a=\u200a7 . 1×10−3 for L1 ) , while the test was inconclusive for L9A/L9a ., Taken together , by using the methods previously developed for studying molecular degeneration associated with suppressed recombination 50 , 51 , we found an elevated genomic degeneration in the mat A chromosomes , which have not been subjected to introgression ., Sex chromosomes , such as the XY in mammals and ZW in birds , and mat chromosome in N . tetrasperma are similar in two main aspects ., Firstly , both harbor a large region of suppressed recombination , which has expanded in a punctuated manner into evolutionary strata 35 , 73 , 74 ., Secondly , both have experienced genomic differentiation and degeneration after the cessation of recombination 50 , 51 , 74 ., However , in terms of introgression , the two sex-determining systems apparently show variable features ., Introgression is unlikely to fixate in most sex chromosomes , since the density of hybrid sterility genes are much higher in sex chromosomes than autosomes , as predicted by Haldanes rule and the large X-effect 11 ., Empirical support for this expectation comes from a wide range of animal taxa including mammals , Drosophila , birds and butterflies 75–77 ., However , in the fungal kingdom , empirical data have shown mixed signals ., Turner et al . ( 2011 ) reported that a large fraction of QTLs for reinforcement of reproductive barriers between heterothallic Neurospora are found on the mat chromosome 78 ., Nevertheless , three previous genealogy studies have found introgression to be enriched in mat chromosome loci compared to autosomal loci 28–30 ., These previous findings together with the large and continuous introgression tracts ( covering >50% of mat chromosome ) reported in this study are suggestive of a relationship between introgression and mating-type loci/chromosomes in fungal species ., The deviation from Haldanes rule and the Large X-effect may originate from the system of determining sexual identity in fungi ., For example , the causes of hybrid sterility in sex chromosomes of animals and plants , including dominance theory , faster-male evolution , faster-X evolution and sex ratio meiotic drive , may not apply to fungal species , which typically exhibit a lack of asymmetry at the mating-type chromosomes and a mixed asexual/hermaphroditic sexual life cycle ., In the absence of the above-mentioned negative effects of introgression occurring on the mating-type loci/chromosomes in fungi , other factors such as genetic invigoration could be the driving forces for the observed association ( see next section ) ., In previous studies we have demonstrated a genetic degeneration in the region of suppressed recombination of the mat chromosomes of N . tetrasperma , both by the accumulation of non-synonymous substitutions and non-preferred synonymous codons 50 , 51 , which fits with theoretical expectations for a genomic region with suppressed recombination 54 ., Notably , Whittle et al . ( 2011 ) reported an asymmetry of degeneration for the two mat chromosomes of N . tetrasperma strain P4492L1 51 and such an asymmetry was later reported by Ellison et al . ( 2011 ) 33 also for strain P581 , belonging to lineage 6 ( L6 38: not included in this study , see Alignment and divergence estimation in materials and methods section ) ., While Whittle et al . ( 2011 ) found an elevated degeneration of mat A in L1 , Ellison et al . ( 2011 ) found an elevated degeneration in mat a of L6 49–51 ., Alternative explanations for an asymmetric degeneration between the mat chromosomes of N . tetrasperma may be proposed , such as uneven level of haploid selection in the two nuclei during the heterokaryotic life cycle 79 , 80 , or introgression ., Here , we confirmed by analyses of both dN/dS and codon usage that the mat A chromosome of L1 ( as well as L4 and L9 ) has experienced more degeneration than the mat a chromosome , and we were able to correlate degeneration to the process of introgression ., Specifically , we hypothesize that the observed introgression of mat a chromosomes in N . tetrasperma from species with free recombination is adaptive , by reducing degeneration levels in this chromosome ., An alternative , neutral , hypothesis is that introgression into N . tetrasperma mat chromosomes is a result of genetic drift ., Currat et al . ( 2008 ) demonstrated through simulations that during a r

Introduction, Results, Discussion, Materials and Methods

The significance of introgression as an evolutionary force shaping natural populations is well established , especially in animal and plant systems ., However , the abundance and size of introgression tracts , and to what degree interspecific gene flow is the result of adaptive processes , are largely unknown ., In this study , we present medium coverage genomic data from species of the filamentous ascomycete Neurospora , and we use comparative genomics to investigate the introgression landscape at the genomic level in this model genus ., We revealed one large introgression tract in each of the three investigated phylogenetic lineages of Neurospora tetrasperma ( sizes of 5 . 6 Mbp , 5 . 2 Mbp , and 4 . 1 Mbp , respectively ) ., The tract is located on the chromosome containing the locus conferring sexual identity , the mating-type ( mat ) chromosome ., The region of introgression is confined to the region of suppressed recombination and is found on one of the two mat chromosomes ( mat a ) ., We used Bayesian concordance analyses to exclude incomplete lineage sorting as the cause for the observed pattern , and multilocus genealogies from additional species of Neurospora show that the introgression likely originates from two closely related , freely recombining , heterothallic species ( N . hispaniola and N . crassa/N . perkinsii ) ., Finally , we investigated patterns of molecular evolution of the mat chromosome in Neurospora , and we show that introgression is correlated with reduced level of molecular degeneration , consistent with a shorter time of recombination suppression ., The chromosome specific ( mat ) and allele specific ( mat a ) introgression reported herein comprise the largest introgression tracts reported to date from natural populations ., Furthermore , our data contradicts theoretical predictions that introgression should be less likely on sex-determining chromosomes ., Taken together , the data presented herein advance our general understanding of introgression as a force shaping eukaryotic genomes .

Introgression is a process by which genetic material from one species becomes infiltrated into another , genetically distinct species ., Introgression usually occurs via sexual reproduction: individuals of two species mate and produce a hybrid offspring , then the offspring repeatedly backcross with one of the parental species ., Introgression has long been recognized as a key process in evolution , as it may contribute to speciation , diversification , and adaptation to new environments ., The importance and prevalence of introgression has been well established in plant and animal systems , and in this study we use a fungal model system , Neurospora , to study the introgression at the genomic level ., We gathered genomic data from six genomes , and by comparative genomics we revealed genetic transfer of DNA regions of unprecedentedly large sizes , covering over 50% of the mating-type chromosomes , and used phylogenetic analyses to reveal the origin and direction of the transfer ., Introgression was found solely on the mating-type chromosomes , which contradicts theoretical predictions for sex-determining chromosomes ., We argue that this unexpected pattern is due to the fact that fungi do not have differentiated sexes ( female/male ) and thereby are free from sex-biased evolutionary forces ., Instead , we suggest that introgression between fungal species may result in reinvigoration of genomic regions exposed to suppressed recombination .

model organisms, genetics, biology, genomics, evolutionary biology, microbiology, genetics and genomics

journal.pcbi.1000390

Computational Models of the Notch Network Elucidate Mechanisms of Context-dependent Signaling

Cells continuously receive signals from their microenvironments – including, factors present in the extracellular matrix , soluble media , and surrounding cells, – which collectively influence cell function and behavior via activating, intracellular signal transduction and gene regulation networks ., These networks, generally involve complex , nonlinear interactions of proteins , such as, phosphorylation cascades ( reviewed in 1 ) and second messenger, signaling systems 2 , whose structures feature positive and negative, feedback loops , feed-forward interactions , signal amplification , and cross-talk with, other pathways 3 ., Mathematical models of these interactions are, therefore very insightful or even necessary avenues to analyze and understand the, regulation of cell behavior , as the properties of these networks can exceed an, intuitive understanding 4–6 ., Notch is a signaling system required for numerous critical cell fate specification, events during the development of the nervous system , hematopoietic system , eye , and, skin 7–11 ., The receptor for, this pathway is the single pass transmembrane protein Notch that , when bound by its, ligands Delta or Jagged , undergoes a series of cleavage events to release its, intracellular domain ( NICD ) 9 , 12 ., This NICD then translocates into the nucleus and acts as a transcriptional, upregulator of target genes , including members of the hes family ,, through its interaction with the transcription factor RBP-Jκ 13 ., In mammals, there are four different Notch proteins ( Notch1-4 ) and 5 ligands ( Delta 1 , 3 , and 4, and Jagged 1 and 2 ) ., For this study , we have focused primarily on the Notch1, signaling pathway ., In its role as a critical regulator of cell fate 7–11 ,, Notch has been known to function via lateral inhibition and induction mechanisms to, create fine-grained patterns in undifferentiated cells , a process required for, proper boundary formation and differentiation of various tissues 14 , 15 ., It, can also function as a binary cell fate switch , for example during differentiation, of the epidermis 16 and endodermal epithelium of the gut 17 , to, promote differentiation of one cell type from precursor cells at the expense of, another ., Furthermore , in some cases continuous Notch activity is not required for, cell fate specification ., For example , transient Delta-Notch signaling has been shown, to be sufficient to induce T-cell 18 and NK cell differentiation 19 from their respective, precursor cells , and can induce an irreversible switch to gliogenesis in neural, crest stem cells 20 ., Notch signaling also occurs only transiently in, many instances during the development of Drosophila, 21 ,, zebrafish 22 , 23 , and mice 24 ., It was also, recently shown that human embryonic stem cells ( hESCs ) require activation of Notch, signaling to form the progeny of all three embryonic germ layers , and subsequent, transient Notch signaling enhanced generation of hematopoietic cells from committed, hESCs 25 ., The mechanisms by which a short Notch signaling pulse can permanently switch cell, fate are not elucidated ., The Notch system has also been shown to function as an oscillator ., Specifically , the, expression levels of members of the hes family , a group of, downstream Notch target genes 26 , have been shown to oscillate with a 2 hour, periodicity in some systems during development , which for example aids in, somitogenesis ( i . e . the patterning of somites ) 27–29 ., Hes1, protein and mRNA concentrations have also been observed to oscillate with an, approximate 2 hr time period upon serum starvation in various cultured cell lines, including myoblasts , fibroblasts , and neuroblastoma cells 30 ., Furthermore ,, oscillations in the Notch network have been proposed to be important in maintaining, neural progenitor cells in an undifferentiated state 31 ., Finally , there is, evidence that such oscillations may also afford cells the opportunity to repeatedly, test for the continued existence of a signal 32 , thereby increasing, cellular response sensitivity and flexibility by allowing the cell to integrate the, results of many periodical evaluations of the signal before making an ultimate cell, fate decision ., The Delta-Notch signaling system has been previously modeled to elucidate its role in, fine-grained pattern formation through the action of lateral inhibition and, induction 33–35 ., Collier et, al . developed a simple 2-parameter model that focuses on pattern formation, due to feedback inhibition between adjacent cells via Delta-Notch signaling 33 ., Other, models build upon this simple model by adding more molecular detail at the, intercellular level 34 , 35 ., In addition , several studies have focused on, trying to understand the underlying mechanism of Notch system oscillations 32 , 36 , where a, Hes1 negative feedback loop composed of Hes1 protein repressing, hes1 transcription , likely plays a central role 37 ., Delays related to transcription and translation were also proposed to be important, for the observed oscillations 38 ., However , while several models have thus been, proposed and have yielded important insights into this system 30 , 36 , 38–40 , they, have focused exclusively on Hes1 and not analyzed its interactions with other, signaling proteins in the Notch system ., Additionally , all these models focus on a, particular aspect or mode of Notch signaling ( e . g . lateral inhibition or, oscillation ) but do not yet address how complex , alternative behaviors could arise, from the same network ., Here we mathematically model the Notch signaling system to analyze how the same, network is capable of functioning as a cell fate switch or an oscillator in, different biological contexts ., This model , which includes the regulation of the, notch1-RBP-Jk-hes1 gene circuit , predicts that the Notch1-Hes1, system acts as a bistable switch in certain regions of parameter space , where Hes1, levels can change by 1–2 orders of magnitude as a function of the input, Delta signal ., In addition , it predicts that a transient pulse of a high level of, Delta is capable of inducing high Hes1 expression levels for a duration that would, be sufficient to induce a cell fate switch ., Moreover , the model elucidates how the, network can be ‘tuned’ to function in different regimes , either, as an oscillator or a cell fate switch , by changing a key parameter ., Finally , low, numbers of reactants can lead to significant statistical fluctuations in molecule, numbers and reaction rates , making cells intrinsically noisy biochemical reactors, 41 , 42 ., Stochastic, simulations of the Notch system , which enable the analysis of the effect of, biological noise in the system arising due to stochastic variations in gene, expression , reveal that for systems that respond quickly to Notch signaling , the, network is able to dampen the effects of this biological noise and function in a, manner similar to what is predicted by the deterministic model ., In summary , the, model enables analysis of the different behavioral responses of the Notch signaling, network observed over a broad spectrum of signaling inputs and parameter values and, can be further expanded to study Notch signaling in numerous contexts ., A set of differential equations was developed to track changes in the, concentrations of various species in the nucleus and cytoplasm of a cell as a, function of time following activation of Notch by its ligand ., The cell is, modeled as a 10 µm diameter sphere with a 5 µm diameter, nucleus ., Numerous processes were modeled as terms in the differential equation, system , including transcription , translation , transport , degradation or - in the, case of Notch - receptor cleavage ( Fig . 1A , Text S1 ) ., As examples , the three equations, tracking the Hes1 cytoplasmic mRNA , Hes1 cytoplasmic protein and Hes1 nuclear, protein concentrations are given by:The rate of change ( in units of ) of the cytoplasmic mRNA concentration of Hes1 is given by the, difference in the rates of it transcription and degradation ., RfHcm is, the transcription rate of hes1 mRNA in the nucleus ., We assume, instantaneous export of mRNA to the cytoplasm ., A factor of 7 is included to take, into account the dilution due to export to the cytoplasm ( Text S1 ) ., kdHcm , kdHcp , and kdHnp denote the, degradation constants for the hes1 mRNA ( Hcm ) , cytoplasmic, protein ( Hcp ) , and nuclear protein ( Hnp ) , respectively , which are assumed to, undergo first order degradation kinetics ., ktrHc denotes the, translation constant ( min−1 ) for conversion of cytoplasmic, hes1 mRNA into cytoplasmic protein ., Transcriptional and, translational delay times are incorporated into the model , as these are, processes that inherently involve delays between initiation and the production, of a molecule of mRNA or protein , as previously described 38 , 44 ., Thus , the translation of Hes1 protein is based on the delayed, hes1 mRNA concentration HcmD ( delayed by time TpHc ,, the average time for translation of Hes1 ) , which is the concentration of mRNA, present when the process of translation was initiated instead of the, concentration at the present time ., kniHcp denotes the nuclear import, rate in units of min−1 ., A dilution factor of 7 is again, used to incorporate differences in nuclear and cytoplasmic volumes ., The transcription rates for notch1 , hes1 , and, RBP-Jκ are based on the states of their respective, promoters ., Previous promoter analysis has been complemented with Genomatix Suite, Gene2Promoter transcription factor ( TF ) binding site prediction software to, identify potential TF binding sites in the promoters of the three genes in the, model ., Takebayashi et al . 37 observed that hes1, transcription is repressed by its own gene product through Hes1 protein binding, to sites in the hes1 promoter termed N-boxes ., Through a series, of binding and transcriptional activity assays , the study determined that Hes1, bound strongly to three N-boxes found upstream of the transcriptional start site, and repressed transcription of the hes1 mRNA up to 40-fold ., Also , while the work concluded that there was a synergistic rather than an, additive effect of the N-box binding dependent repression of gene expression ,, further mathematical analysis has indicated that there is no or very weak, synergy among the different binding sites 45 ., Several positive, regulatory regions were also found in the hes1 promoter , and it, was also shown to have two adjacent RBP-Jκ binding sites 46 , 47 ., Thus , we have, modeled the hes1 promoter to have three equivalent N-boxes, where the Hes1 protein can bind and repress transcription , as well as two, equivalent RBP-Jκ sites ., The presence of all other positive regulators, of transcription is lumped into a constant basal rate of transcription ., As it has not been extensively investigated , the notch1 promoter, sequence was analyzed in the Gene2Promoter software ., One putative Hes1 site, ( N-box ) and two putative RBP-Jκ sites were found in the ∼1 kb, notch1 promoter analyzed ., This may imply that, notch1 is both positively and negatively regulated by its, own gene product ., To test this , a transcriptional activity experiment using the, dual luciferase assay system was conducted ., The promoter of murine, notch1, 48, was used to drive expression of hRluc cDNA ( Renilla luciferase ) ., Co-transfection, studies with plasmids expressing RBP-Jκ , NICD , Hes1 , and dNHes1 ( a, dominant negative form of Hes1 ) indeed demonstrated that the, notch1 promoter is regulated negatively by Hes1 and, RBP-Jκ in the absence of NICD but is positively regulated by NICD in the, presence of RBP-Jκ ( Text S1 , Fig . S1 ) ., The notch1 promoter was modeled with two RBP-Jκ sites, and one N-box ., A 418 bp sequence upstream of the RBP-Jκ gene as, characterized by Amakawa et al . 49 was analyzed in, the Gene2Promoter software for TF binding sites of interest ., Three potential, Hes1 binding sites and three potential RBP-Jκ sites were found ., Thus ,, the RBP-Jκ gene also potentially undergoes, autoregulation under Notch signaling , and a three N-box , three RBP-Jκ, site model was utilized ., As discussed above , all three promoters have one or more binding sites for both, Hes1 ( the N-box ) and RBP-Jκ ., It is assumed that Hes1 can bind and, repress transcription of the corresponding promoter only in its homodimer form ,, and the dimerization reaction is assumed to be at steady state over timescales, of protein transcription , translation and import , driven by mass action kinetics, such that the concentration of the dimer is given by:Where , KaHp is the association equilibrium constant, for the dimerization reaction ., Similarly , the time scales of transcription, factor binding to and dissociation from the promoter elements are also assumed, to be much faster than those of gene transcription and protein synthesis , such, that binding to the promoter is at pseudo steady state ., In addition , it is, assumed that NICD can bind only when an RBP-Jk protein is bound to its site on, the promoter , and that this NICD binding converts RBP-Jκ from a, transcriptional repressor to an activator 13 ., The level of promoter activation ( i . e . rate of mRNA synthesis ) is modeled by an, approach termed BEWARE 50 , 51 , in which the probabilities of a promoter, being in any one of its many possible states are calculated based on the, relative concentrations of the three transcription factors ( Hes1 , RBP-Jκ, and NICD ) , and their respective DNA binding affinities , using equilibrium, binding equations ., The level of activation of the promoter is then given by: , where , PPi is the probability, of the promoter being in state i , and vi is the activation rate of, gene transcription associated to the promoter being in that state i ., When the, promoter is empty , the gene activation rate is assumed to be the basal, transcription rate ( Vb ) for that promoter ., When a Hes1 dimer is bound, to an N-box , the rate is reduced by a factor rN that takes into, account the repressive effect of the Hes1 transcription factor , and when, RBP-Jκ is bound , the rate is reduced by a factor rR ., Furthermore , when the promoter is in its maximally activated state with the NICD, bound to the RBP-Jκ and no Hes1 dimers bound , the activation rate is, assumed to be at its maximum and is given by, ( Vmax+Vb ) ., In the case of multiple RBP-Jκ, binding sites , an additional factor tc ( <1 ) is used to account, for states where not all RBP-Jκ sites bind NICD to represent the, decrease from the maximum possible activation rate ., For a detailed expression of, transcription rates please refer to Supplemental Materials ( Text S1 ) ., Although explicit parameters have been included to account for cooperative, binding for Hes1 dimers to multiple N-boxes and for RBP-Jκ binding, ( cooperativity factors Cn , Cr and Cnr - please, refer to Table 1 for model, parameters ) , they have been set to 1 for these simulations , as recent work, suggests there is very little if any cooperative effect in Hes1 binding to, N-boxes 45 ., Finally , it is assumed that each mRNA produces a, fixed number of proteins , i . e . mRNA dynamics have been neglected 50 ., Experimentally determined values for half-lives of proteins and mRNA , association, and dissociation constants of proteins to their respective DNA binding sites ,, dimerization constants , and protein translation and transcription rates have, been used when possible ( Table, 1 ) ., These values are often not available for the exact species of, interest; however , the best available estimates based on similar protein classes, are used wherever applicable as the starting point ., The time delays for, transcription and translation for each of the three genes are calculated as, previously described 52 and are detailed in the Supplemental, Materials ( Text, S1 ) ., 4 . 5 transcripts per minute 45 and 20 transcripts, per minute 53 were used as initial estimates for, hes1 basal and maximum transcription rates respectively ., The transcription rates for RBP-Jκ and, notch1 were then determined from these estimates and the, estimates of their minimum transcription times ( Text S1 ) ., The degradation rates for the Hes1 protein and mRNA were determined, experimentally by Hirata et al . in fibroblasts 30 ., They observed, similar values in other cultured cell types including myoblasts , neuroblastomas ,, and teratocarcinomas ., Pulse chase experiments of Logeat et al . 54, were used to assess the degradation rates for the full-length Notch1 protein ,, and an estimate of Notch1 protein half-life of ∼40 minutes was derived ., GSK3β has been shown to affect the stability of NICD 55 ., Although there are conflicting results as to whether GSK3β helps to, stabilize 55 or destabilize the cleaved NICD 56 ,, our experimental results show that GSK3β is essential for the NICD, regulation of neural stem cell differentiation into astrocytes ( Agrawal , Ngai ,, and Schaffer , manuscript in preparation ) ., Furthermore , we show that Notch1, signaling upregulates the expression of GSK3β in these cells ., Thus , the, effect of GSK3β is incorporated into the model by increasing the, half-life of NICD from 3 to 8 hrs 55 above a threshold, concentration of Hes1 ( which is assumed to directly or indirectly regulate the, expression of GSK3β ) ., This increased NICD half-life does not however, change the qualitative behavior of the Hes1 switch ( Fig . S3A ) ., The repression constant of Hes1 dimer bound to an N-box ( rNbox ) is, estimated from the results of Takebayashi et al . 37 that show that, in the presence of three N-boxes , transcription is repressed by ∼40, fold ., This yields a repression value of ∼0 . 3 per N-box ( Please refer to, Supplemental Materials ( Text S1 ) for details ) ., Since there are no, reliable estimates of the NICD generation constant upon Delta binding, ( kfNcp ) , a lumped parameter of this constant with the Delta, concentration is used to report the strength of the Delta signal, ( kfNcp*Delp ) ., The initial parameters for which the, experimentally determined values are not accurately available were later, subjected to sensitivity analysis ( See results ) ., The differential equations described in the model were solved ( with parameter, values given in Table 1 ), using Berkeley Madonna 8 . 3 . 11 software ( www . berkeleymadonna . com ) with the Runge Kutta 4 module at a step, size of 1 min ., To arrive at realistic initial conditions for the model , the, initial concentrations of all species were set to 0 with zero Delta signal , and, the simulations were run until the various species attained steady state, concentration levels ., These steady state values ( listed in Table 2 ) were then used as, the initial conditions for subsequent simulations ., For the various experiments ,, the system was run for 750 minutes without stimulation with the Delta ligand to, attain a basal steady state , and the Delta concentration was then increased to, different levels to initiate Notch1 signaling ., Simulations were run either with, a constant Delta signal throughout or with varying duration pulses of the Delta, signal ., The system was simulated for a duration of 5 , 000–10 , 000, minutes ( ∼3 . 5–7 days ) , as neural progenitor stem cells have, been previously shown to undergo differentiation upon Notch activation in, 3–5 days ( 57 ) ., Longer simulations up to 50 , 000 minutes, were conducted when required to confirm Hes1 had reached steady state levels ., Since the levels of several protein species in the deterministic model, simulations were very low ( Table, 2 ) , at the level of tens of molecules per cell , assumptions of mass, action kinetics and pseudo steady state may not hold true , and stochastic, effects may play an important role in the dynamics of the signaling network, 41 , 58 ., To analyze whether, noise in protein and mRNA concentrations would impact the dynamics of the, system , a stochastic simulation of the model using the Gillespie algorithm 43, was implemented in C++ ( code available upon request ) ., To relax, the assumptions of mass action kinetics and pseudo steady state , we explicitly, simulated every reaction step , making a total of 299 reactions ., For example ,, every interaction between a transcription factor and a promoter was modeled as a, discrete reaction in the simulation ., The τ-leap method 59, was also incorporated into the algorithm to accelerate the stochastic, simulations and increase their efficiency ., The response of the Notch1-RBP-Jκ-Hes1 system to a step change in an, input Delta signal was analyzed ., Simulations were initiated using the steady, state levels of the different species in the absence of any external Delta ( also, listed in Table 2 ) , and at, t\u200a=\u200a750 minutes a Delta signal was applied ., Fig . 2A demonstrates, that when a low input Delta stimulus is applied , the Hes1 concentration settles, to a correspondingly low steady state value ., However , when the input Delta, signal was increased ( 10-fold ) , Hes1 shows a rapid increase to a new , 20-fold, higher steady state value ., Further steady state analysis at a range of input, Delta levels and initial conditions reveals that the system exhibits, bistability ., At low levels of Delta signal , basal levels of Hes1 are maintained, in the cell ( “OFF” state ) , but as the Delta signal strength, is increased beyond a threshold level , it stimulates the production of Hes1 ,, which is then maintained at high levels ( “ON” state ) through, the concerted regulation of the Notch1-RBP-Jκ-Hes1 network ( Fig . 2B ) ., Bistability, – which has previously been proposed as an advantageous mechanism to, mediate an unambiguous cell fate switch , including in stem cells 51 , 60, – is evident within an intermediate range of Delta signal values, ( Fig . 2B ) ., The initial numbers of some protein and mRNA species in the system were in the, range of tens of molecules per cell ( Table 2 ) , such that stochastic fluctuations, in individual species may impact the dynamics of the network ., In particular ,, intracellular noise inherent in systems with small numbers of molecules and/or, slow biochemical reactions can randomize or undermine the, “accuracy” of cell fate choices 41 , 58 ., To analyze such, behavior , stochastic simulations based on the Gillespie algorithm 43 ,, distinct from the deterministic model , were developed ., Steady state analysis, shows that at low , constant Delta signals , the Hes1 levels fluctuate about a low, mean value corresponding to the “OFF” state , as expected, ( data not shown ) ., However , if the Delta signal is increased to a level just, below the concentration at which the deterministic model would predict a switch, in state ( Fig . 2B ) ,, stochastic simulations reveal that noise in the network can induce some, trajectories to spontaneously switch states ( Fig . 3A ) ., Analogous to results previously, observed in other systems 51 , 61 , 62 ,, noise thus undermines the bistable switch and induces spontaneous flipping, between states ., Analysis of the time it takes the system to initially pass from, the lower to the upper state reveals that as the strength of the input signal is, increased , this average first passage time ( FPT ) decreases , and the percentage, of trajectories that change state increases ( Fig . 3B ) ., However , this, “uncertainty” occurs within a narrow range of intermediate, Delta signal levels , and if this intermediate window is avoided , the system, effectively behaves deterministically ., In addition , “ON” to “OFF” transitions, were simulated by first stimulating with a high Delta signal for 4000 minutes to, induce high Hes1 expression levels ., When Delta was then reduced to levels that, were in the predicted bistable region based on the deterministic model , the, system maintained high expression levels of Hes1 ( Fig . 4A ) , as anticipated from the, deterministic results ( Fig ., 2B ) ., Contrary to what was expected based on the deterministic model ,, however , when the Delta signal was instead reduced to zero , some trajectories, remained in the high Hes1 expression ( “ON” ) state ( Fig . 4B ) ., This indicates the, role of stochastics in potentiating high Hes1 expression levels even in the, absence of continued signal ., It has been shown for neural crest stem cells 20 that a transient, Notch signal is sufficient to induce cell differentiation ., Also , there are, numerous situations where transient Notch-Delta signaling determines the fates, of immature cells , both in tissue culture 18 , 19, and during organismal development 21–24 ., Under continuous Delta stimulation , the system, can attain high steady-state Hes1 expression levels , thus acting as a switch ,, but we next wanted to examine whether transient Delta activation was also, capable of eliciting high Hes1 expression ., We thus examined the dynamic response, of the system to transient activation of the Notch1 pathway upon variation in, the strength and duration of an applied Delta signal ., When the system is stimulated for a short duration ( 10 minutes ) with a moderate, strength Delta signal , the deterministic model predicts a transient peak in the, Hes1 expression that eventually decays to its low steady state value ( Fig . 5A ) ., However , the peak, expression of Hes1 continually increases with increasing input signal duration, up to ∼800 minutes , beyond which the maximum expression levels of Hes1, attained remain the same but the duration of prolonged high expression levels, progressively increases ( Fig ., 5A ) ., Similarly , as the input Delta signal strength is increased for a, constant pulse duration , the peak Hes1 concentrations attained also increase up, to a maximum value , after which a further increase in the signal strength only, increases the duration of high Hes1 levels ( Fig . 5B ) ., The cell is thus able to attain, high Hes1 expression either under prolonged low intensity Delta signaling or a, short burst of high intensity Delta signaling ., We also examined the effect of stochastics on transient activation of the, network ., Simulations were run using the parameter values as in the deterministic, model for various Delta pulse durations ranging from 10 minutes to 3000 minutes, and >40 trajectories per input duration value were analyzed ., For Delta, pulse durations of less than 500 minutes , the stochastic simulations followed, the prediction of the deterministic model ( data not shown ) ., However , for a, 500-minute Delta pulse , even though the deterministic model predicts a transient, Hes1 peak that does not attain the maximum possible expression level , a small, percentage of the stochastic trajectories in fact did switch to the, “ON” state ( corresponding to high Hes1 expression levels ), ( data not shown ) ., Also , as the duration of the Delta pulse is increased , the, percentage of trajectories that remain in the “ON” state for, the simulated 15 , 000 minutes progressively increases even though the, deterministic model predicts that the system would revert back to the, “OFF” state within that time ., Furthermore , the average first, passage time ( FPT ) of the trajectories that do switch state increases as the, Delta pulse duration increases ( Fig . 6 ) ., It is likely that for shorter Delta pulse durations , if the, system is to undergo the spontaneous “OFF” to, “ON” transition , it does so early , soon after the, application of the Delta signal ., However , in the case of longer duration input, signals , the continued presence of the signal allows trajectories to switch, state even much later in the simulation , resulting in an apparently longer first, passage time ., Collectively , these results imply that even for very short signal, pulse , a small fraction of a population of cells receiving a pulse of Delta, signal could switch their state due to stochastic effects ., A number of parameters in the model have not been directly experimentally, measured and were estimated from data available for similar protein classes in, different contexts , and we thus performed sensitivity analysis for all such, parameters by varying them individually through a broad range of values in the, deterministic model ( Table, 3 , Fig ., S2 ) ., Although in most cases the qualitative behavior of the system, remained unchanged , the system did exhibit considerable sensitivity to specific, parameters , which were then subjected to further analysis ., These include: the, half-life of NICD , the equilibrium binding constant of NICD with RBP-Jκ, ( Ka ) , the maximal transcription rates ( Vmax ) , and the, repression constant of Hes1 ( rNbox ) ., NICD has a long half-life of a, few hours under normal physiological conditions 55 ., However , our model, indicates that if the NICD half-life is drastically reduced , the system fails to, function as a switch and cannot express high levels of Hes1 ( Fig . S3 ) ., In addition , the equilibrium binding constant ( Ka ) of NICD to, RBP-Jκ in the model is 108 M−1 , but as, Ka increases – denoting stronger interactions of NICD, with the promoter – bifurcation analysis demonstrates that the OFF-ON, transition occurs at accordingly lower values of the Delta signal, ( kDelp ) ( Fig ., 7A ) ., Similarly , increasing the maximal transcription rate of Hes1, ( Vmaxh ) to indicate a stronger promoter shifts the OFF-ON, transitions to lower Delta signal strengths ( Fig . 7B ) ., Interestingly , the response of the deterministic model was most sensitive to the, extent to which Hes1 binding reduced or repressed expression of target genes, ( rNbox ) ., As the Hes1 repression constant ( rNbox ) is, progressively decreased ( or the repressive strength of Hes1 progressively, increased ) from 0 . 3 to 0 . 1 , the final steady state concentrations of Hes1, progressively decrease for a given level of Delta signaling ( Fig . S4 ) ,, but the system continues to exhibit bistability ., Intriguingly , as the value of, rNbox is further decreased below 0 . 1 , there is a dramatic, qualitative change in the response of the system ., Specifically , the system, undergoes a bifurcation or transition from bistable to monostable behavior and, at such high repressive strengths is unable to attain high steady state Hes1, expression levels ., Finally at very low values of rNbox, ( <0 . 03 ) , it once again undergoes a transition to a stable oscillatory, response where the Hes1 levels in the cell oscillate about a low mean steady, state value ( Fig . 8A ) ., A, phase plot of the response of the system with variable rNbox ( Fig . 8B ) demonstrates how the, same gene network can transition from behaving as a bistable switch to being an, oscillator ., The model thus elucidates the versatility of the system , where, tuning of a single key parameter can convert its behavior from a switch to a, clock ., Previous hes1 models showing sustained oscillations have, focused exclusively on the low rNbox region ( i . e ., rNbox\u200a=\u200a0 ) of such a phase plot, 30 , 36 , 38 , 39 ., The Notch signaling system is an evolutionarily conserved network that functions in, multiple organs to orchestrate cell fate specification 63–65 in a, context dependent manner ., In s

Introduction, Methods, Results, Discussion

The Notch signaling pathway controls numerous cell fate decisions during, development and adulthood through diverse mechanisms ., Thus , whereas it functions, as an oscillator during somitogenesis , it can mediate an all-or-none cell fate, switch to influence pattern formation in various tissues during development ., Furthermore , while in some contexts continuous Notch signaling is required , in, others a transient Notch signal is sufficient to influence cell fate decisions ., However , the signaling mechanisms that underlie these diverse behaviors in, different cellular contexts have not been understood ., Notch1, along with two downstream transcription factors hes1 and, RBP-Jk forms an intricate network of positive and negative, feedback loops , and we have implemented a systems biology approach to, computationally study this gene regulation network ., Our results indicate that, the system exhibits bistability and is capable of switching states at a critical, level of Notch signaling initiated by its ligand Delta in a particular range of, parameter values ., In this mode , transient activation of Delta is also capable of, inducing prolonged high expression of Hes1 , mimicking the, “ON” state depending on the intensity and duration of the, signal ., Furthermore , this system is highly sensitive to certain model parameters, and can transition from functioning as a bistable switch to an oscillator by, tuning a single parameter value ., This parameter , the transcriptional repression, constant of hes1 , can thus qualitatively govern the behavior of, the signaling network ., In addition , we find that the system is able to dampen, and reduce the effects of biological noise that arise from stochastic effects in, gene expression for systems that respond quickly to Notch signaling ., This work thus helps our understanding of an important cell fate control system, and begins to elucidate how this context dependent signaling system can be, modulated in different cellular settings to exhibit entirely different, behaviors .

The Notch signaling pathway is an evolutionarily conserved signaling system that, is involved in various cell fate decisions , both during development of an, organism and during adulthood ., While the same core circuit functions in various, different cellular contexts , it has experimentally been shown to elicit varied, behaviors and responses ., On the one hand , it functions as a cellular oscillator, critical for somitogenesis , whereas in other situations , it can function as a, cell fate switch to pattern developing tissue , for example in the, Drosophila eye ., Furthermore , malfunctioning of Notch, signaling is implicated in various cancers ., To better understand the underlying, mechanisms that allow the network to function distinctly in different contexts ,, we have mathematically modeled the behavior of the Notch network , encompassing, the Notch gene along with two of its downstream effector transcription factors ,, which together form a network of positive and negative feedback loops ., Our, results indicate that the qualitative and quantitative behavior of the system, can readily be tuned based on key parameters to reflect its multiple roles ., Furthermore , our results provide insights into alterations in the signaling, system that lead to malfunction and hence disease , which could be used to, identify potential drug targets for therapy .

cell biology/cell signaling, developmental biology/cell differentiation, computational biology/signaling networks

journal.pgen.1000803

Postnatal Survival of Mice with Maternal Duplication of Distal Chromosome 7 Induced by a Igf2/H19 Imprinting Control Region Lacking Insulator Function

Parthenogenetic mouse embryos usually die before 6½ days post coitum ( dpc ) ., Occasionally they develop to the 25 somite forelimb bud stage or approximately 9½ dpc 1–5 ., Parthenogenones possess two maternally-derived genomes and would be expected to possess abnormal levels of transcript of all known imprinted genes , that is , lack of expression of paternally expressed genes ( two inactive copies ) , and over-expression of maternally expressed genes ( two active copies ) ., Their death is likely a composite effect of at least some of these misexpressions , although those involved are not well defined ., Defining the causes is important for improving understanding of the aetiology of genomic imprinting 6–9 and the prevalence of sexual reproduction , which ‘has long been an evolutionary enigma’ 10 ., Knowledge of the causes of parthenogenetic death has come from two sources ., First , the union of unbalanced complementary gametes in intercrosses of mice carrying reciprocal or Robertsonian translocations yield , at low frequency , embryos with maternal duplication and paternal deficiency for particular Chr regions as defined by the translocation breakpoint 11–13 ., Maternal duplication of twelve Chr regions results in developmental anomalies ., Only three of these are associated with peri- or prenatal death , these being maternal duplication of proximal Chr 6 ( MatDup . prox6 ) —prior to 11½ dpc 14 , maternal duplication of distal Chr 7 ( MatDup . dist7 ) —late fetal death 15 , and maternal disomy of Chr 12—perinatal death , probably attributable to the distal region 16 ., Second , knockouts of imprinted genes and imprinting control regions ( ICRs ) have provided information on the effects of disregulation of imprinted genes , for example , 17–21 ., To better define the causes of failed parthenogenetic development , and learn more of how imprinted genes at dist7 work together to regulate normal development , we have examined some of the misexpressions of imprinted genes thought to contribute to the abnormal development of MatDup . dist7 conceptuses ., These display a pronounced growth deficit of the fetus and placenta and die at the late fetal stage , or possibly at birth ., Live MatDup . dist7 young have never been observed 13 , 15 ( J . Mann , unpublished data ) ., Dist7 is an important region in terms of genomic imprinting , containing over 20 imprinted genes 13 , 22 ., At least three of these are regulated by the Igf2/H19 imprinting control region ( ICR ) , these being ‘insulin like growth factor 2’ ( Igf2 ) —paternally expressed and encoding a mitogen important for embryonic growth 23 , 24 , ‘insulin II’ ( Ins2 ) —paternally expressed in yolk sac 25 , and the non-coding ‘H19 fetal liver mRNA’ ( H19 ) gene—maternally expressed 26 ., Other non-coding transcripts have been described , these being Mir483 , contained within an intron of Igf2 27 and for which imprinting status is unknown , Mir675 , contained with an H19 exon and therefore likely to follow the imprinting pattern of the host gene 28 , 29 , and antisense transcripts within Igf2 30 ., The targets of the Mir483 and Mir675 miRNAs are unknown ., The maternally-derived Igf2 allele is inactive due to the hypo-methylated maternal Igf2/H19 ICR functioning as a‘CCCTC-binding factor’ ( CTCF ) -based chromatin insulator ., This lies between the Igf2 promoter and the shared Igf2-H19 enhancers , preventing their interaction ., The maternal H19 promoter lies on the same side of the insulator as the enhancers , therefore interaction occurs ., On the paternal Chr the ICR is hyper-methylated , preventing CTCF binding and insulator formation and allowing for paternal Igf2 promoter and enhancer interaction ., The paternal H19 promoter , just distal to the methylated ICR , also becomes methylated , and is inactive ., The Ins2 gene is located just distal to Igf2 ., The Ins2 parental alleles are affected in the same way as their Igf2 counterparts , but only in yolk sac ., Ins2 is expressed biallelically in pancreas 25 , 31–33 ., Telomeric or distal to the Igf2/H19 ICR domain is a large cluster of imprinted genes under regulatory control of the Kv differentially methylated region ( DMR ) -1 ( KvDMR1 ) ICR ., The active state of maternally-derived genes within this cluster is coincident with maternal-specific ICR methylation and the inactive state of the promoter of the ‘KCNQ1 overlapping transcript 1’ ( Kcnq1ot1 ) gene contained within the ICR ., The paternal ICR is hypo-methylated , and paternal-specific elongation of the Kcnq1ot1 transcript is coincident with silencing in cis of genes within the cluster 17 , 34 , 35 ., One of the genes regulated by this ICR is the ‘cyclin-dependent kinase inhibitor 1C ( P57 ) ’ ( Cdkn1c ) gene encoding a protein facilitating reduced cell proliferation , increased apoptosis and delayed cell differentiation 36 , 37 ., MatDup . dist7 fetuses are maternally duplicated for the hypo-methylated Igf2/H19 ICR and hyper-methylated KvDMR1 ICR regions , as well as for other imprinted transcripts at dist7 ., This epigenetic configuration is highly similar to that associated with the human imprinting-related growth deficit disorder , Silver-Russell syndrome ( SRS ) ( OMIM 180860 ) ., More than half of cases are associated with hypo-methylation of the IGF2/H19 ICR , also known as ‘ICR1’ ., The disease is also associated with maternal duplication of the KvDMR1 ICR region , also known as ‘ICR2’ , and maternal duplication of the 11p15 . 5 Chr region encompassing both ICRs ., It is strongly suspected that SRS is caused by downregulation of IGF2 , and , in a minority of cases , excess CDKN1C or other imprinted genes regulated by ICR2 ., However , empirical evidence is lacking 38–40 ., The death of MatDup . dist7 fetuses has been difficult to decipher ., Available evidence suggests that maternal duplication of the Igf2/H19 ICR regulatory domain alone is insufficient to explain the total phenotype observed ., Mice with paternal inheritance of a tandem duplication of a chicken β-globin CTCF-based chromatin insulator , substituted for the endogenous Igf2/H19 ICR , are similar to MatDup . dist7 mice in having a fully functional hypo-methylated insulator on both parental Chrs ., They lack Igf2 activity , have at least twofold over-expression of H19 , with both parental alleles probably active , and would be expected to lack Ins2 activity in yolk sac ., Nevertheless , their phenotype—dwarfism combined with postnatal viability—is essentially identical to Igf2 mutants 41 ., Mice homozygous for this genetic modification , in a mix of strains 129S1/SvImJ and outbred Swiss CF-1 , showed normal fecundity and were maintained as a random-bred line for several years ( J . Mann , unpublished data ) ., Further , lack of Igf2 activity is unlikely to be the sole cause of reduced growth in MatDup . dist7 fetuses ., At 17½ dpc , their weight is approximately 40% of wild-type 42 ( J . Mann and Walter Tsark , unpublished observations ) compared to 50–60% of wild-type for Igf2 mutants and mice maternally inheriting the chicken insulator 41 ., Overall , these observations indicate that the MatDup . dist7 phenotype of fetal growth deficit and death involves the misexpression of imprinted genes outside the influence of the Igf2/H19 ICR , and this has previously been suggested 42 ., Available evidence also indicates that maternal duplication of the KvDMR1 ICR regulatory domain alone is insufficient to explain the total phenotype observed ., Mice with paternal inheritance of a deletion of this element exhibit biallelic expression of adjacent imprinted genes ., These mice , in a mix of mouse strains 129S4/SvJae and C57BL/6J , are postnatally viable ., They show some reduction in size , and it has been indicated that this is caused by over-expression of Cdkn1c 35 ., Reduced growth has also been observed in Cdkn1c-BAC transgenic mice ., While these displayed high frequency perinatal mortality in strain 129/Sv , high postnatal viability was obtained in a mix of strains 129/Sv and outbred Swiss MF1 43 ., These observations indicate that MatDup . dist7 late fetal death , occurring in the context of mixed strains including outbred Swiss , involves the misexpression of imprinted genes outside the influence of the KvDMR1 ICR ., Overall , these observations have led to suggestions that MatDup . dist7 death could be a composite effect of misexpressions derived from both imprinted domains , for example , Igf2 inactivity combined with Cdkn1c over-expression 43 ., To define the role of imprinted genes regulated by the Igf2/H19 ICR in the MatDup . dist7 phenotype , we evaluated the effects of introducing a mutated Igf2/H19 ICR ( ICRΔ ) which cannot bind CTCF and form an insulator 44 ., MatDup . dist7 fetuses carrying ICRΔ would be expected to be corrected in terms of the number of active alleles of Igf2—activation of one of two inactive alleles , H19—repression of one of two active alleles , and Ins2—activation of one of two inactive alleles in yolk sac ., MatDup . dist7 fetuses carrying ICRΔ were significantly rescued in terms of growth and were able to survive to adulthood ., These results demonstrate that the aberrant phenotype of MatDup . dist7 fetuses is highly dependent on the presence of two maternally-derived Igf2/H19 ICR chromatin insulators ., Maternal inheritance of ICRΔ results in activation of Igf2 in cis such that total Igf2 RNA is 1 . 7 and 2 . 1 times the normal level in the liver and kidney of 17½ dpc fetuses , respectively , and also repression of H19 in cis , such that total H19 RNA is 0 . 2 and 0 times the normal level in these same tissues , respectively 44 ., This configuration of expression—two active Igf2 and two inactive H19 alleles—is coincident with increased growth , an effect thought to be due to the former misexpression 18 , 45 , 46 ., Lack of H19 RNA alone has no effect on Igf2 expression or imprinting and results in no discernible phenotype 47 ., Maternal inheritance of ICRΔ would also be expected to result in activation of Ins2 in yolk sac ., To confirm that maternal inheritance of ICRΔ can mediate normal growth , we tested its function in mice paternally inheriting a null mutation of Igf2 ( Igf2− ) ., Mice of genotype ( ICR+/+ , Igf2+/− ) are small due to lack of Igf2 activity , with the maternal allele inactive , and the paternal allele null 24 ., Results are shown in Figure 1 ., Experimental young of genotype ( ICRΔ/+ , Igf2+/− ) , in which the maternally-derived Igf2 allele is activated in cis by ICRΔ , were not significantly different in weight to control ( ICR+/+ , Igf2+/+ ) mice at 6 weeks of age ( females , P\u200a=\u200a0 . 271; males , P\u200a=\u200a0 . 035 ) ., Thus , a single maternal copy of ICRΔ induces sufficient Igf2 activity for achieving normal postnatal growth ., We note that , in respect to growth with one versus two active Igf2 alleles , experimental ( ICRΔ/+ , Igf2+/− ) animals with one active allele ( maternal ) , were not significantly different in weight to ( ICRΔ/+ , Igf2+/+ ) animals with two active alleles ( females , P\u200a=\u200a0 . 378; males , P\u200a=\u200a0 . 089 ) ., Further , ( ICR+/+ , Igf2+/+ ) females with one active allele ( paternal ) , were not significantly different in weight to ( ICRΔ/+ , Igf2+/+ ) females with two active alleles ( P\u200a=\u200a0 . 04 ) ., However , in males , mice with one active allele ( paternal ) were lighter than mice with two active alleles , as expected ( P\u200a=\u200a0 . 002 ) ., Given the borderline probability values obtained , greater numbers of animals need to be analysed to accurately determine the relative growth rates of mice of the various genotypes ., MatDup . dist7 zygotes were produced in intercrosses of mice carrying the reciprocal translocation T ( 7;15 ) 9H ( T9H ) ., Such intercrosses give rise to a high proportion of unbalanced zygotes , and litter size is small ., Of balanced zygotes , only one in seven are expected to be MatDup . dist7 , these identified by the dist7 marker , albino ( c ) , a mutation of the ‘tyrosinase’ ( Tyr ) gene 15 ., The ICRΔ mutation was introduced into female T9H/+ parents and was inherited by MatDup . dist7 zygotes ( Figure 2 ) ., Expected allelic activity of Igf2 and Cdkn1c in the three possible MatDup . dist7 genotypes is shown ( Figure 2B ) ., ICRΔ-induced activation of Igf2 was confirmed in 13½ dpc MatDup . dist7 fetuses obtained in ( T9H/+ , Tyrc/c , ICRΔ/+ ♀×T9H/+ , Tyr+/+ , ICR+/+ ♂ ) intercrosses ., The level of Igf2 transcript in MatDup . dist7 ICRΔ ., + fetuses was the same as in control ICR+/+ fetuses with one active allele , while it was almost double the normal amount in MatDup . dist7 ICRΔ ., Δ fetuses with probably two active alleles ( Figure 3A and 3B ) ., Increased total Igf2 RNA was also seen in mice which maternally inherit ICRΔ and have an active maternal and paternal allele of Igf2 ( Figure 3A and 3B ) ., Also , MatDup . dist7 fetuses of all genotypes contained at least double the amount of Cdkn1c RNA relative to controls , probably because of two active alleles ( Figure 3A and 3B ) ., These intercross matings were allowed to proceed to term and we immediately began to observe viable albino or MatDup . dist7 young which were of overtly similar size to agouti littermates ., A MatDup . dist7 animal and its two littermates at 10 days post-partum is shown ( Figure 4 ) ., All MatDup . dist7 young obtained were of genotype ICRΔ ., + or recombinant ICRΔ ., Δ ., Seven of 52 mice born were MatDup . dist7 which is similar to the expected frequency , indicating that ICRΔ was always able to increase growth and rescue viability ., In age- and litter-matched animals , a significant weight deficit of approximately 17% in MatDup . dist7 animals became apparent at 6 weeks of age when compared with controls carrying an equivalent number of active Igf2 alleles , that is , MatDup . dist7 ICRΔ ., + with control ICR+/+ ( one active allele each ) and MatDup . dist7 ICRΔ ., Δ recombinant with control ICRΔ/+ ( probably two active alleles each ) ( Figure 5A ) ., CDKN1C may antagonize the growth promoting effects of IGF2 17 , 48 , and it has been suggested that excess CDKN1C may combine with lack of IGF2 to cause MatDup . dist7 death 43 ., To test this possibility , we introduced a null allele of Cdkn1c ( Cdkn1c− ) into MatDup . dist7 fetuses to enforce its monoallelic expression ., In ( T9H/+ , Tyrc/c , ICR+/+ Cdkn1c+/− ♀×T9H/+ , Tyr+/+ , ICR+/+ Cdkn1c+/+ ♂ ) matings , all of 55 young obtained were agouti controls , that is , at least six albino MatDup . dist7 ( ICR+ . + , Cdkn1c+ . − ) pups were expected , but none were observed ., This result is not consistent with the idea that MatDup . dist7 death results only from the combined action of the Cdkn1c and Igf2 misexpressions ., To test for a role of Cdkn1c over-expresssion in the growth deficit at 6 weeks of age of rescued postnatal MatDup . dist7 ICRΔ ., + animals , we introduced Cdkn1c− into MatDup . dist7 fetuses such that they were of genotype ( ICRΔ . + , Cdkn1c+ . − ) ., This genotype should be normalized for the number of active alleles of imprinted genes regulated by the Igf2/H19 ICR , and also be normalized for Cdkn1c expression , that is , all of these imprinted genes should be monoallelically expressed ., In ( T9H/+ , Tyrc/c , ICRΔ/+ , Cdkn1c+/− ♀×T9H/+ , Tyr+/+ , ICR+/+ , Cdkn1c+/+ ♂ ) matings , viable MatDup . dist7 ICRΔ ., + , Cdkn1c+ ., − young were obtained and these did not display a significant weight deficit at 6 weeks of age—with the caveat that the weight measurements are relative to control young obtained in the previous matings ( Figure 5B ) ., Their weights could not be compared to littermates as , given the mating scheme , agouti littermates were always positive for ICRΔ—inheritance of Cdkn1c− being lethal—and therefore possessed two active copies of Igf2 ., In any event , these results are consistent with the possibility that biallelic expression of Cdkn1c does contribute to a reduction in postnatal growth in MatDup . dist7 ICRΔ ., + or ICRΔ ., Δ , Cdkn1c+ ., + animals ., We have shown that maternal introduction of a mutant Igf2/H19 ICR , which lacks chromatin insulator activity , into MatDup . dist7 fetuses substantially alters their abnormal phenotype—small size and death at the late fetal stage—to one of near normal growth rate and survival to adulthood ., This result clearly demonstrates the dependence of this phenotype on a misexpression of imprinted genes caused by the presence of two active maternally-derived Igf2/H19 ICR chromatin insulators ., As this ICR is known to regulate the expression of at least three dist7 imprinted genes—H19 , Ins2 , Igf2 , and a number of non-coding transcripts—correction in the misexpression of one or more of these was probably responsible for the result obtained ., Activation of Igf2 was likely an important correction , this being the only alteration in expression induced by ICRΔ expected to affect growth ., The survival of MatDup . dist7 mice with ICRΔ is more difficult to decipher ., As discussed in the Introduction section , it is unlikely that the Igf2/H19 ICR-derived misexpressions are solely responsible for their death , as mice with two functional chromatin insulators—a maternally-derived Igf2/H19 ICR , and a paternally-derived chicken insulator substituted for the Igf2/H19 ICR , possess the same combination of misexpressions as MatDup . dist7 mice in respect to this region , yet these animals have normal postnatal viability 41 ., Further evidence is provided by observations of the effects of misexpression of each imprinted gene alone ., First , for H19 , no overt effect on phenotype is observed in transgenic mice with ectopic over-expression 49–52 ., Biallelic or over-expression of H19 has been suggested to cause perinatal death of mice produced by combining a non-growing oocyte genome ( ng ) , carrying a deletion of the distal Chr 12 IG-DMR ICR ( Δ12 ) , with a fully grown oocyte genome ( fg ) —ngΔ12/fg mice 53 ., However , these mice would be predicted to have the equivalent expression profile of imprinted genes as mice with maternal inheritance of the chicken insulator substitution ., The latter mice are viable , despite twofold over-expression of H19 41 ., Therefore , the perinatal death of ngΔ12/fg mice may result from the combined action of H19 RNA excess—or possibly Igf2 RNA absence—and small imperfections in expression derived from the non-growing oocyte genome , for example , as related to the IG-DMR ICR deletion ., Second , for Ins2 , mice lacking in expression of this gene are viable 54 ., Third , for Igf2 , mice lacking expression are dwarfed and have impaired lung development 55 , but are usually viable ., High postnatal survival frequency of Igf2 mutants is seen in inbred strain 129/SvEv 23 , 24 although in this strain we have observed a low level of perinatal death ( J . Mann , unpublished observations ) ., In the present study , in a mix of strains 129/SvEv and outbred Swiss CF-1 , we observed high frequency survival ., Also , in this same strain mix , we maintained a Igf2−/− random-bred line for a number of years which had normal fecundity ( J . Mann , unpublished data ) ., On the other hand , use of a second Igf2 null mutation 56 revealed that lack of IGF2 in strain C57BL/6J results in death at birth ., This effect was not peculiar to this second knockout allele as homozygous mutants can be obtained in strain 129 ( M . Constancia , personal communication ) ., In the present study , MatDup . dist7 young were a mix of strains 129S1/SvImJ , CF-1 , C57BL/6J and CBA/Ca ., In this mix , lack of Igf2 activity is highly likely to be compatible with survival ., Given these various lines of evidence , the present experiments strongly suggest that misexpressed imprinted genes , as regulated by the Igf2/H19 ICR , work in combination with misexpressions derived outside of this region of influence in causing the total MatDup . dist7 phenotype ., The significant rescue in growth probably mediated by Igf2 activation may also be directly related to MatDup . dist7 survival in that it could compensate for negative effects derived from outside the Igf2/H19 ICR region ., Nevertheless , we cannot rule out the possibility that Ins2 inactivity in yolk sac , excess H19 RNA , or the misexpression of non-coding RNAs regulated by the Igf2/H19 ICR make a contribution to the lethal effect ., These possibilities could be investigated through correction of their misexpression in MatDup . dist7 fetuses , then determining growth and survival ., For example , correction of H19 over-expression could be achieved by introducing a deletion of the transcript region only ., The imprinted genes operating outside the influence of the Igf2/H19 ICR that contribute to MatDup . dist7 death would be expected to require maternal- , rather than paternal-specific imprinting or methylation for attaining differential expression in the normal context ., This is because for full-term development , there is apparently no other requirement , aside from Igf2/H19 ICR methylation , for paternal imprinting at dist7 57 ., The cluster of genes requiring maternal-specific methylation of the KvDMR1 ICR for activity fulfills this criterion ., While the introduction of a null mutation of Cdkn1c , and hence enforced monoallelic expression of this gene , did not rescue MatDup . dist7 fetuses , this does not rule out the possibility that CDKN1C excess has a role in causing MatDup . dist7 death ., In MatDup . dist7 ( ICR+ . + , Cdkn1c+ . + ) fetuses , Cdkn1c RNA levels were found to be more than three times that of controls , suggesting that each maternally-derived Cdkn1c allele was upregulated 1 . 5-fold ., Therefore , CDKN1C could still be in excess in MatDup . dist7 ( ICR+ . + , Cdkn1c+ . − ) animals ., Also , there remains the possibility that excess Cdkn1c RNA may contribute as part of a network of misexpressions derived from the cluster regulated by the KvDMR1 ICR ., For example , biallelic expression of the ‘pleckstrin homology-like domain , familiy A , member 2’ ( Phlda2 ) gene results in placental growth retardation and marginal fetal growth restriction 58 , and upregulation of PHLDA2 is correlated with growth retardation in humans 59 , 60 ., Also , it has been suggested that excess expression of the ‘achaete-scute complex homolog 2 ( Drosophila ) ( Ascl2 ) gene could cause the MatDup . dist7 lethal effect 42 . The phenotype of MatDup . dist7 fetuses could also involve misexpressions of dist7 imprinted genes lying outside of the influence of the two known ICRs . For example , ‘adenosine monophosphate deaminase 3’ ( Ampd3 ) —maternally expressed in placenta , and identified in a transcriptome analysis of MatDup . dist7 conceptuses 22 , ‘inositol polyphosphate-5-phosphatase F’ ( Innp5f ) —an isoform paternally expressed in brain 61 , and ‘cathepsin D’ ( Ctsd ) —possible paternal-specific expression 62 ., The postnatal weight deficit of approximately 17% in MatDup . dist7 young at 6 weeks of age was similar to that in mice paternally inheriting a deleted KvDMR1 ICR ., This deletion results in biallelic expression of imprinted genes regulated by this ICR , including Cdkn1c 17 , 34 ., Indeed , excess CDKN1C has been indicated as the cause of the weight deficit 35 ., Consistent with this possibility is that the weight of MatDup . dist7 ICRΔ ., + , Cdkn1c+ ., − young was normal at 6 weeks of age ., However , we note that MatDup . dist7 neonates displayed no significant weight deficit until reaching adulthood , while in mice paternally inheriting the deleted KvDMR1 ICR , the weight deficit is present in fetuses and persists throughout postnatal development 17 ., More data regarding weight gain in relation to the inheritance of ICRΔ , in MatDup . dist7 young and otherwise , is required to confirm these observations ., In terms of MatDup . dist7 death , additional experiments are required to determine exactly which combination of misexpressions are involved ., The total MatDup . dist7 phenotype has been ascribed to the very distal portion of Chr 7 as defined by the reciprocal translocation T ( 7;11 ) 65H ( T65H ) 42 ., This translocation has a breakpoint far more distal on Chr 7 relative to the T9H translocation used in this study , although is still proximal to the two clusters of imprinted genes regulated by the Igf2/H19 and KvDMR1 ICRs ., However , some caution should be exercised in ascribing the total effect to this region ., While it was shown that T65H- and T9H-MatDup . dist7 fetuses are of similar morphology 42 , the postnatal viability of the former was not investigated ., If T65H-MatDup . dist7 fetuses are also inviable , then the composite lethal effect is likely to be contained within the two aforementioned clusters of imprinted genes ., Evidence that the KvDMR1 cluster contributes to the effect could be obtained by determining the viability of MatDup . dist7 fetuses carrying a deletion of this whole cluster ., This would result in enforced monoallelic expression of all genes under regulation of the KvDMR1 ICR , including Cdkn1c , and these mice and would be expected to be postnatally viable , although small because of Igf2 inactivity ., Such a deletion , made through truncation of Chr 7 at a point distal to the Ins2 gene , has been described 63 ., A complication with this possible experiment is the existence of imprinted genes at dist7 which are not regulated by either ICR ., Another experiment could be to breed mice with paternal inheritance of the chicken β-globin insulator substituted for the Igf2/H19 ICR 41 combined with paternal inheritance of the KvDMR1 ICR deletion 17 ., These would misexpress all imprinted genes under regulatory control of both ICRs ., If these were the only misexpressions involved in the MatDup . dist7 phenotype , then the phenotype should be reproduced ., MatDup . dist7 fetuses provide an epigenetic model of a subtype of human Silver-Russell syndrome ( SRS ) involving maternal duplication of the orthologous Chr region , 11p15 . 5 , which encompasses ICR1 and ICR2 ., In these fetuses , we have shown that abrogation of ICR1 insulator function was able to restore Igf2 expression , concomitant with restoration of growth and survival ., The most common subtype of SRS , that involving hypo-methylation of ICR1 , is perhaps better modelled in mice maternally inheriting the chicken insulator in place of ICR1 ., These animals provide information on the effects of the presence of two functional insulators at the Igf2/H19 region as the only epigenetic lesion ., In these fetuses , we previously showed that DNA methylation was abrogated while insulator function remained intact ., This resulted in reduced Igf2 activity and growth retardation 41 ., Both of these findings support the idea that reduced expression of IGF2 during fetal development is causal in the development of SRS ., They also support the suggestion that the failure to detect low concentrations of serum IGF2 in SRS patients is related to downregulation of IGF2 by this stage 38 ., Further genetic manipulation in these mouse models should provide additional implications for the human disease ., Our experiments suggest that misexpression of imprinted genes caused by two maternal copies of the Igf2/H19 ICR constitute one component of a composite barrier to parthenogenetic development that was not previously predicted ., The lethal effect in MatDup . dist7 fetuses may be specific to later stages of development , and may not normally occur in parthenogenones given their peri-implantation death ., Nevertheless , high-level paternal- and maternal-specific expression of Igf2 and H19 , respectively , is present shortly after implantation , at least by 6½ dpc 64 ., Therefore , it cannot be ruled out that these misexpressions , and others regulated by the Igf2/H19 ICR , play a role in what probably is a complex composite lethal effect involving a network of misexpressed imprinted genes ., Indeed , the fact that parthenogenones fail earlier in development than embryos with maternal duplication of any single Chr region , indicates that misexpressions of imprinted genes from different regions are cumulative or synergistic in their deleterious effects ., Further , at the molecular level , it has been shown that disregulation of the imprinted genes ‘pleiomorphic adenoma gene-like 1’ ( Plagl1 ) and H19 can affect the expression of other imprinted genes in an imprinted gene expression network 65 , 66 ., Previous observations have shown that the normal activity of imprinted genes regulated by the Igf2/H19 ICR are one of a small number of developmentally critical expression profiles provided exclusively by imprinting through the male germ line , provided that most if not all other imprinted genes are not misexpressed 57 ., The present results raise the possibility that full-term parthenogenetic development could be achieved by correcting the misexpressions of only a few imprinted genes in order to repair the total expression network ., One necessary correction would be to activate the ‘paternally expressed 10’ ( Peg10 ) gene ., Lack of expression of this gene results in death by 10½ dpc , and this misexpression alone would be expected to present a barrier to parthenogenesis ., It would be expected to contribute to , or could be solely responsible for , the embryonic death of MatDup . prox6 mice , which occurs prior to 11½ dpc 20 ., Line no ., ; genotype; strain; source , how produced , or reference: Line-1; 129S1/SvImJ ( 129S1 ) ; Tyr+/+; The Jackson Laboratory , stock no . 002448 ., Line-2; outbred Swiss CF-1; Tyrc/c; Charles River Laboratories ., Line-3; T9H/T9H , Tyr+/+; mix of C57BL/6J ( B6 ) and CBA/Ca ( CB ) ; The Jackson Laboratory , stock no . 001752 ., Line-4; T9H/T9H , Tyrc/c; mix of B6 , CB and CF-1; made by mating line-2 with -3 , then intercrossing ., Line-5; Tyrc/c , ICRΔ/Δ; mix of CF-1 and 129S1; made by mating ICRΔ/+ mice 44 with line-2 , then intercrossing ., Line-6; Tyrc/c , Cdkn1c+/−; mix of 129S7/SvEvBrd ( 129S7 ) , B6 and CF-1; made by mating Cdkn1c+/− mice 37 with line-2 , then intercrossing ., Line-7; T9H/T9H , Tyrc/c , Cdkn1c+/−; mix of strains B6 , CB , CF-1 , and 129S7; made by mating line-4 with -6 , then intercrossing ., Line-8; Igf2+/−; 129/SvEv 23 ., Production of experimental ( ICRΔ/+ , Igf2+/− ) mice ( Figure 1 ) : Female parents ( ICRΔ/+ , Igf2+/+ ) were bred in ( line-5 ♀×line-1 ♂ ) matings ., Male parents ( ICR+/+ , Igf2+/− ) were of line-8 ., Young were a mix of strains 129 and CF-1 ., Production of MatDup . dist7 ICRΔ ., + and ICRΔ ., Δ mice ( Figure 2 ) : Female parents ( T9H/+ , Tyrc/c , ICRΔ/+ ) were bred in ( line-5 ♀×line-4 ♂ ) matings ., Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings ., Young were a mix of strains 129S1 , CF-1 , B6 , and CB ., Production of MatDup . dist7 Cdkn1c− ., + young , attempted: Female parents ( T9H/+ , Tyrc/c , Cdkn1c+/− ) were bred in ( line-4 ♀×line-6 ♂ ) matings ., Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings ., Young were a mix of strains 129 , B6 , CB , and CF-1 ., Production of MatDup . dist7 ( ICRΔ . + , Cdkn1c+ . − ) young ( Figure 5B ) : Female parents ( T9H/+ , Tyrc/c , ICRΔ/+ , Cdkn1c+/− ) were bred in ( line-5 ♀×line-7 ♂ ) matings ., Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings ., Young were a mix of strains 129S1 , 129S7 , B6 , CB , and CF-1 ., For the ICR , two pairs of primers were used ., The first pair was specific for the mutant ICR , identical to a pair previously described 41: 5′- GCCC ACCA GCTG CTAG CCATC -3′ and 5′- CCTA GAGA ATTC GAGG GACC TAAT AAC -3′ , 240 bp amplicon identified ICRΔ ., + and ICRΔ ., Δ animals ., The second pair was specific for ICR+ , with primers binding to sequence positions that were modified in ICRΔ 44: 5′- AACA AGGG AACG GATG CTAC CG -3′ and 5′- GCAA TATG TAGT ATTG TACT GCCA CCAC -3′ , lack of a 506 bp amplicon identified ICRΔ ., Δ animals ., For Cdkn1c , the null allele was identified using primers specific for

Introduction, Results, Discussion, Materials and Methods

The misexpressed imprinted genes causing developmental failure of mouse parthenogenones are poorly defined ., To obtain further insight , we investigated misexpressions that could cause the pronounced growth deficiency and death of fetuses with maternal duplication of distal chromosome ( Chr ) 7 ( MatDup . dist7 ) ., Their small size could involve inactivity of Igf2 , encoding a growth factor , with some contribution by over-expression of Cdkn1c , encoding a negative growth regulator ., Mice lacking Igf2 expression are usually viable , and MatDup . dist7 death has been attributed to the misexpression of Cdkn1c or other imprinted genes ., To examine the role of misexpressions determined by two maternal copies of the Igf2/H19 imprinting control region ( ICR ) —a chromatin insulator , we introduced a mutant ICR ( ICRΔ ) into MatDup . dist7 fetuses ., This activated Igf2 , with correction of H19 expression and other imprinted transcripts expected ., Substantial growth enhancement and full postnatal viability was obtained , demonstrating that the aberrant MatDup . dist7 phenotype is highly dependent on the presence of two unmethylated maternal Igf2/H19 ICRs ., Activation of Igf2 is likely the predominant correction that rescued growth and viability ., Further experiments involved the introduction of a null allele of Cdkn1c to alleviate its over-expression ., Results were not consistent with the possibility that this misexpression alone , or in combination with Igf2 inactivity , mediates MatDup . dist7 death ., Rather , a network of misexpressions derived from dist7 is probably involved ., Our results are consistent with the idea that reduced expression of IGF2 plays a role in the aetiology of the human imprinting-related growth-deficit disorder , Silver-Russell syndrome .

Parthenogenetic mouse embryos with two maternal genomes die early in development due to the misexpression of imprinted genes ., To gain further insight into which misexpressions might be involved , we examined some of the misexpressions that could determine the small size and fetal death of a “partial parthenogenone”—embryos with maternal duplication of distal Chr 7 ( MatDup . dist7 ) ., We investigated the involvement of two maternal copies of the Igf2/H19 imprinting control region ( ICR ) , which is associated with lack of activity of the Igf2 gene , encoding a growth factor , and over-activity of H19 ., By introducing a mutant ICR , we activated Igf2 and expected to correct other misexpressions , such as that of H19 ., The result was substantial increase in growth and full postnatal viability of MatDup . dist7 fetuses , demonstrating the dependency of their abnormal phenotype on two maternal copies of the ICR ., Activation of Igf2 was probably the main effector of this rescue ., These results are consistent with the idea that reduced expression of IGF2 is causal in the human growth deficit disorder , Silver-Russell syndrome .

genetics and genomics/gene expression, developmental biology/developmental evolution, genetics and genomics/epigenetics, developmental biology/developmental molecular mechanisms, evolutionary biology/developmental evolution

journal.pgen.1000392

Gene Dosage Effects of the Imprinted Delta-Like Homologue 1 (Dlk1/Pref1) in Development: Implications for the Evolution of Imprinting

Genomic imprinting , the process causing genes to be differentially expressed according to their parental origin acts on a subset of developmentally regulated genes reviewed in 1 ., Since imprinting results in functional haploidy of target genes , the evolution of such a mechanism must have conferred a significant advantage to its recipients to offset the cost ., In addition , the maintenance of this method of gene dosage regulation must confer a continuing selective benefit to the individual ., By creating genetic models that mimic the loss of imprinting of genes regulated in this way , we can study phenotype and consider what selective pressures act to maintain gene dosage , and begin to examine what may have driven the acquisition of mono-allelic expression ., Dlk1 , Delta-like homologue 1 also known as Preadipocyte factor 1 ( Pref1 ) , on mouse chromosome 12 is part of an imprinted gene cluster 2 , 3 ., It encodes a transmembrane glycoprotein that possesses six epidermal growth factor-like motifs in the extracellular domain similar to those present in the Delta/Notch/Serrate family of signalling molecules ., In contrast to other NOTCH ligands , DLK1 does not have the Delta:Serrate:Lin-12 ( DSL ) domain believed to mediate the interaction and activation of the NOTCH receptor 4 ., Nevertheless , DLK1 can interact with NOTCH through specific EGF-like repeats and can act as a Notch antagonist , both in culture and in vivo , binding to the receptor without activating it 5 , 6 , 7 ., In vitro , Dlk1 maintains precursor cell populations and inhibits differentiation 4 , 8 , 9 ., Dlk1 is expressed at high levels in a wide range of embryonic tissues 10 , 11 however its developmental functions in vivo are largely unknown ., In more ancestral vertebrates such as the fish Oryzias latipes and Fugu rubripes , Dlk1 , and its neighbour Dio3 which encodes a negative regulator of thyroid hormone metabolism , are only 10–15 Kb apart ., In eutherian mammals , the genes have become separated by the insertion of sequences including a retrotransposon-like gene ( Rtl1 ) and several non-coding RNAs , including a large microRNA cluster , and have acquired imprinting 12 ., Dlk1 , Rtl1 and Dio3 are expressed from the paternally inherited chromosome and the non-coding RNAs from the maternally-inherited chromosome 13 ., Imprinting on chromosome 12 is controlled by an intergenic differentially methylated region ( IG-DMR ) that is methylated on the paternally inherited chromosome 14 , 15 ., The unmethylated maternal IG-DMR is necessary to repress protein-coding transcription and to activate the non-coding RNAs 14 ., In terms of gene expression , the paternal chromosome 12 most likely resembles the ancestral ( pre-imprinted ) state as Dlk1 and Dio3 are expressed and the non-coding RNAs are silenced 12 ., Furthermore , paternal deletion of the methylated IG-DMR has no effect on transcription 14 ., Alteration of imprinting at chromosome 12 has considerable consequences for embryonic fitness ., Uniparental disomy mice with two paternal copies and no maternal copy ( PatDi ( 12 ) /PatDp ( dist12 ) ) or mice with two maternal copies and no paternal copy ( MatDi ( 12 ) /MatDp ( dist12 ) ) of chromosome 12 , are lethal and show distinct phenotypes ., These include growth abnormalities and developmental defects in muscle , cartilage/bone and placenta 16 , 17 , 18 ., Animals die prenatally commencing at E16 ., The embryonic uniparental disomic phenotypes can be ascribed to the Dlk1-Dio3 imprinted cluster because embryos with maternal deletion of the IG-DMR ( ΔIG-DMRMAT ) recapitulate the transcriptional profile and embryonic mutant phenotypes of the PatDi ( 12 ) /PatDp ( dist12 ) conceptuses 14 , 19 ., Since loss of imprinting at chromosome 12 is lethal , there are clear advantages in maintaining the imprinted state ., However since in PatDi ( 12 ) /PatDp ( dist12 ) and ΔIG-DMRMAT conceptuses multiple paternally expressed coding genes are up-regulated , and multiple maternally expressed non-coding genes are repressed , we are unable to determine if selection is acting to maintain the mono-allelic expression of one or all protein coding genes or to maintain expression of the non-coding RNAs ., As Dlk1 is an ancestral gene at this imprinted locus and it is expressed at high levels in tissues where uniparental disomy conceptuses have the most pronounced abnormalities 10 , we consider Dlk1 as a strong candidate for the pathological defects associated with the Dlk1-Dio3 region ., Therefore in this study , we specifically ask if imprinting might be maintained to prevent overdose of Dlk1 and moreover , if selection to control the dosage of this gene might have driven the evolution of imprinting of the chromosome 12 cluster ., Manipulating imprinted gene dosage in vivo provides a powerful tool to study imprinted gene function and evolution ., Here we describe a Dlk1 over-expression model in which the gene is driven by its own endogenous regulatory sequences ., We generate mouse lines harbouring bacterial artificial chromosome ( BAC ) transgenes that encompass the entire unmanipulated Dlk1 gene with different lengths of flanking sequence ., Of these , transgenic Dlk1 expression at sites where the endogenous gene is expressed , was achieved from a 70 kb BAC transgene in three independent lines and the outcome was the same for all three lines ., This allowed us to compare the phenotypes of mice expressing a double ( transgene hemizygotes ) or triple ( transgene homozygotes ) dose of Dlk1 with normal animals expressing a single imprinted dose ., This transgenic system of regulated Dlk1 over-dose allowed us to explore the developmental and physiological functions of this gene in comparison with other models , and infer evolutionary scenarios for Dlk1 imprinting ., BAC transgenes were generated that encompass the entire Dlk1 gene and endogenous flanking sequences but without other genes in the cluster ., Transgenic mice were created by pronuclear injection of two BAC transgenes differing in the amounts of flanking regulatory sequences ( Figure 1A ) : TgDlk1-31 starts 8 kb upstream of the Dlk1 gene and ends approximately 18 kb downstream of the Dlk1 transcriptional start site ., The TgDlk1-70 transgene shares its 3′ end with the TgDlk1-31 transgene but contains 49 . 4 kb of sequence upstream of Dlk1 ( Figure 1A ) ., The four independent lines containing the TgDlk1-31 transgene failed to express Dlk1 and no Dlk1-associated phenotypes were observed ( data not shown ) regardless of transgene copy number ( Figure S1A ) ., These mice were not analysed further ., In contrast , the TgDlk1-70 transgenic animals successfully expressed Dlk1 ( Figure 1B and C and Table S1 ) , and were subjected to further analysis ., The four independent TgDlk1-70 lines of mice , hitherto referred to as 70A , 70B , 70C and 70D were maintained as hemizygotes ( referred to as WT/TG ) and bred to C57BL/6 for more than ten generations to ensure stability of copy number and phenotype ., Copy number was stable from the third generation onwards and estimated to be 4–5 ( 70A ) , 5–6 ( 70B ) , 7 ( 70C ) and 1 ( 70D ) ( Figure S1 and Table S1 ) ., Once stable lines were established , analysis of the overall level of Dlk1 expression was performed by Northern Blotting for the different lines ., In the single copy 70D line , no transgene-derived Dlk1 expression was observed and no phenotypic consequences were noted ( data not shown ) ., This line was not analysed further ., Importantly , for the 70A , 70B and 70C lines , WT/TG E16 embryos expressed approximately twice as much Dlk1 as their normal littermates regardless of copy number ., Representative quantitative Northern blot expression data is illustrated in Figure 1B and 1C for 70C and 70B families ., Detailed expression data ( both from Northern blots and RT-qPCR ) are shown independently for the three lines in Figure S2B and STable S1 ) ., This expression analysis using two independent methods clearly shows that WT/TG E16 and E18 embryos for the three lines express approximately twice as much Dlk1 ., Interestingly , expression of the transgene occurred independent of the parental-origin of the transgene in the three independent lines ( Figure 1B; Table S1 ) ., Absence of transgene imprinting was confirmed using a single nucleotide polymorphism ( SNP ) located in the 3′UTR of Dlk1 ( Lin et al . , 2003 ) allowing the endogenous gene to be distinguished from the transgene ( Figure S2A; Table S1 ) ., As the transgene is not imprinted , homozygous transgenic mice ( TG/TG ) were generated to further increase the dosage of Dlk1 in the three transgene expressing lines ., As illustrated in Figure 1C for 70B and shown also for 70A and 70C ( Figure S2 and Table S1 ) levels of Dlk1 in the E16–E18 TG/TG embryos were approximately 3-fold higher than normal littermates ., Dlk1 regulation is complex involving alternative splicing and extensive post-translational modifications , therefore we compared DLK1 protein isoforms between the genotypes ( for the 70B family ) ., DLK1 protein levels in the different genotypes are consistent with the Dlk1 mRNA expression and we saw no change in isoform preference ( Figure 1D; Figure 2B ) ., Furthermore , DLK1 protein levels in WT/TG embryos are comparable to that of PatDi ( 12 ) embryos , and are further increased in TG/TG conceptuses ., Overall levels of Dlk1 expression were assessed in WT/WT and WT/TG embryonic tissues at E16 ( Table S1 ) and E18 by TaqMan RT-qPCR for 70B ( Figure 2A ) and 70A ( Table S1 ) ., Results show that Dlk1 levels are over-expressed around 2–2 . 5 fold in embryonic tissues analysed in the WT/TG when compared to normal littermates ( Figure 2A ) ., The placenta was the only tissue with no significant over-expression ( 109%×±15% compared to WT/WT ) ( Figure 2A and Table S1 ) ., This suggests that key placenta specific regulators lie outside the region delineated by the transgene ., Western Blot analysis of DLK1 protein expression was conducted in control , WT/TG and TG/TG tissues and compared with the RT-qPCR analysis ., In all cases , using this semi-quantitative method , protein levels correlated with RNA levels for the tissues and stages analysed ( Figure 2B ) ., Curiously , a slight increase in DLK1 protein levels was observed in the TG/TG placenta , suggesting that there is minimal expression of the transgene in this tissue ., Tissue-specific expression of Dlk1 was compared between normal and transgenic conceptuses by in situ hybridisation and no ectopic Dlk1 activity was evident ( Figure 3 and Table S4 ) at E16 or E18 ., This suggests that endogenous regulators are present on the transgene and are driving transgenic Dlk1 transcription ., Tissue-specific expression of the endogenous neighbouring non-coding RNA gene , Gtl2 was unaffected in all genotypes ( data not shown ) allowing phenotypes of the transgenic mice to be attributed to Dlk1 over-expression ., Phenotypic characterization was performed independently in all three lines and all the observed phenotypes were consistent so data was combined ., Results generated from the individual lines are presented and also summarised in the Supplementary Data ., Imprinted genes have long been known to function in controlling pre-natal growth and nutrient acquisition 20 ., To address a growth function for Dlk1 , we performed extensive measurements of wet and dry embryonic masses , placental wet masses and crown-rump lengths ., WT/TG fetuses are expressing a double dose of Dlk1 and showed consistent overgrowth from E16 to the day of birth ., Wet and dry embryonic masses and crown-rump ( C–R ) length values increased by 6–10% compared to WT/WT littermates ( Figure 4A–B; Table S3 ) ., TG/TG fetuses show growth enhancement at E16 ( wet mass and dry mass increases of 23–25% ) ., In contrast , no differences are observed in E18 C–R length and dry mass compared to WT/WT littermates most likely due to the failure to thrive of these severely compromised fetuses at the later stages as has been observed in other models 16 ., Differences in wet mass between TG/TG and WT/WT at late gestation are due to oedema ( Figure 6A–B ) ., This late gestation data correlates with lethality of the TG/TG fetuses commencing around E16 ( Table 1 , Figure 4 and see below ) ., To determine the contribution of individual tissues to the increase in fetal mass , E18–19 brains , livers , lungs , forelimbs , hindlimbs and brown adipose tissue ( BAT ) were weighed and analysed histologically ( Figure 4C , Figure 5 , Figure 6 ) ., In the growth-enhanced WT/TG fetuses , lungs , forelimbs and hindlimbs were proportionally heavier ., WT/TG brains , that significantly over-express Dlk1 , were spared ., E18 liver growth was unaffected , suggesting that the milder Dlk1 over-expression in this tissue ( E18: 1 . 31×±0 . 21 of WT/WT Dlk1 liver expression and at E16: 1 . 67×±0 . 21 ) may not be sufficient to cause overgrowth ., Overgrowth was observed in TG/TG livers evident from E16 ., Histological examination revealed no obvious defects and portal triads , and hepatocyte size appeared unaffected ( Figure 4C , Figure 6 and data not shown ) ., In general , this suggests that , with the exception of the brain , multi-organ overgrowth is associated with Dlk1 over-expression in an organ-autonomous manner ., In the placentas , no significant differences between the genotypes were observed at any stage ( Figure 4A and Table S3 ) ., Histological analyses indicated no obvious phenotypic abnormalities in the placenta at E16 and E19 for the three genotypes ( data not shown ) ., This is consistent with the finding that Dlk1 is expressed at very low levels from the transgene in this tissue ( Figure 2 and Table S1 ) and suggests that embryonic growth enhancement occurs independently of the placenta ., PatDi ( 12 ) /PatDp ( dist12 ) and ΔIG-DMRMAT exhibit costal cartilage defects and hypo-ossification of mesoderm-derived bones 16 , 18 , 19 ., Since WT/TG embryos have increased crown-rump length , we examined the skeletal development of WT/TG and TG/TG mice ., WT/TG skeletons show signs of growth enhancement and minor ossification delays in the sternum and the closure of the sagittal suture ( Figure 5 ) ., Further increasing the dosage of Dlk1 in TG/TG animals led to more severe skeletal defects ., Overall the TG/TG skeleton was smaller and the delay in the closure of the sagittal suture was more pronounced ., In addition , we observed a bell-shaped thorax with a hypo-ossified sternum and thinner ribs and vertebrae ., We can therefore relate the severity of hypo-ossification with increasing levels of Dlk1 ., Another major defect associated with PatDi ( 12 ) /PatDp ( dist12 ) and ΔIG-DMRMAT is skeletal muscle immaturity 16 , 18 , 19 ., Interestingly , in contrast to the defects in the skeleton , no muscle maturation abnormalities were observed even when Dlk1 dosage was tripled ., Detailed morphometric analysis assessing myofiber diameter and % of myofibers with centrally located nuclei was performed on the E18 diaphragm and on a subset of muscles in the forearm and no differences were observed ( Figure S3 ) ., Dlk1 over-expression of all isoforms was confirmed at the protein level ( Figure 2B ) ., Therefore our data clearly show no prenatal muscle immaturity or hypertrophy induced by Dlk1 over-expression from endogenous regulators ., Hemizygous transgenic embryos are viable and fertile and observed at Mendelian frequencies at birth indicating that a double dose of Dlk1 expression is compatible with embryonic viability ( Table 1 and . Table S2 ) ., In contrast to WT/TG animals , TG/TG embryos have distinct morphological features from E16 including severe oedema , a small thoracic region and a protruding abdomen ( Figure 6A and B ) ., This is associated with lethality from E16 with none of the TG/TG animals surviving more than a few hours after birth ( Table 1 and Table S2 ) ., We also observed that the weight of the TG/TG lungs was significantly reduced ( Figure 4C ) with a denser cellular arrangement at E18 ( Figure 6C ) ., Dlk1 is strongly expressed in the bronchioles of the lung ( Figure 3 and ref 10 ) and an intrinsic defect in lung development caused by over-expression of Dlk1 is likely to contribute to the lethality of these animals ., Despite the absence of major embryonic abnormalities caused by doubling Dlk1 dosage in vivo , we observed a significant increase in early postnatal lethality in WT/TG animals ., 32% of WT/TG pups died during the first three days after birth compared to 10% of WT/WT littermates , in crosses from wild type mothers ( Table 1 and Table S2 ) ., In order to understand the cause of death , we studied processes essential to early postnatal survival in rodents such as temperature regulation , suckling ability and glucose homeostasis ., Temperature regulation is required for adapting to cold exposure after birth and relies on the process of non-shivering thermogenesis ( NST ) mediated by brown adipose tissue ( BAT ) metabolism ., Dlk1 is believed to inhibit adipogenesis of both brown and white adipose tissue 21 , 22 thus is a likely candidate for involvement in the NST process ., We measured BAT per body weight and analysed markers for both fat differentiation ( such as Pparγ2 ) and thermogenesis ( such as Ucp1 ) before and after birth ., At E19 , WT/TG BAT per body weight and levels of Pparγ2 and Ucp1 are comparable to WT/WT ( Figure S4 ) ., At birth , we observed that Ucp1 and Pparγ2 expression was slightly elevated , perhaps to compensate for a slight decrease in BAT mass per body weight ( Figure 7A and B ) ., Failure to thrive , therefore , cannot be ascribed to compromised NST ., The suckling ability of Dlk1 WT/TG pups at birth was assessed by using stomach weight as a measure of milk content 23 ., We observed that WT/WT animals were frequently found with stomach content in the range of 0 . 03–0 . 08 grams , whilst WT/TG animals never exceeded 0 . 03 grams ( Figure 7A ) ., In situ data shows that at late gestation ( E18 ) , highest expression of Dlk1 is in the tongue and the upper and lower lips , which is consistent with a role of Dlk1 in suckling ( Table S4 and data not shown ) ., Blood glucose levels were not different at birth between the two genotypes ( Figure 7C ) ., We also decided to monitor the growth performance of WT/TG animals during the first three weeks of life , when juveniles are still dependent on the mother for nutrition ., WT/TG animals are born bigger than their littermates ( Figure 4A and Figure 7A ) , but they fail to gain weight at the same rate as WT/WT neonates during the first two weeks ( Figure 7D ) ., By 14 days , WT/TG animals are small and remain small thereafter ( ∼10% ) ( Figure 7D and data not shown ) ., The differences in growth performance are most pronounced within the first week after birth ( Figure 7D ) , correlating with poor suckling and the increased lethality of these animals ., In conclusion , despite an early growth advantage , animals expressing a double dose of Dlk1 fail to thrive in early life , and thus any benefit conferred by an increased embryonic size is offset by postnatal lethality ., We have developed a unique model in which the consequences of a single , double and triple dosage of one imprinted gene , Dlk1 , can be assessed in the growing embryo ., The double dose is reminiscent of the situation where there is no imprinting of this gene ., BAC and YAC transgenes have been widely used in mouse genetics and in imprinting studies as a molecular tool for the localisation of transcriptional or imprinting regulatory elements 24 , 25 , 26 ., We generated transgenes differing in the extent of the sequences upstream of Dlk1 ., The TgDlk1-31 did not express Dlk1 in four transgenic lines , regardless of integration site or copy number ( Figure S1A ) ., In contrast , TgDlk1-70 , which extends from 49 kb upstream of the Dlk1 gene , is expressed ., We have therefore determined that the majority of embryonic tissue-specific enhancer sequences for Dlk1 expression are located in the 8 kb to 49 kb interval upstream of the gene ., This is consistent with a previous report showing that enhancers for Dlk1 are absent from 3 kb upstream to 175 kb downstream of the gene 26 ., Minimal transgene expression in the placenta suggests the absence of the specific regulatory sequences for this organ in the 70 kb transgene ., At E16 , we detected growth enhancement in a Dlk1 dose-dependent manner ., TG/TG were more than 20% larger than WT/WT littermates , whereas WT/TG were growth-enhanced by 10% ., At later stages , progressive morbidity and mortality of TG/TG embryos precluded a meaningful comparative assessment of their growth rate ., This was not the case for viable WT/TG embryos which clearly maintained an increased growth trajectory from E16 up to birth ., This is reciprocal to the 20% reduction in weight reported for Dlk1-null E19 fetuses 21 ., In general , tissue-specific overgrowth correlated with levels of Dlk1 over-expression , except in the brain ., This suggests Dlk1 acts locally on organ growth , though we cannot rule out an endocrine role 27 ., With the exception of studies on placental specific Igf2 28 , 29 , previous functional analyses of imprinted genes expressed in embryo and placenta have not been able to discern whether growth effects are intrinsic to the embryo or are secondary to placental function ., This is because genetic manipulation of imprinted genes often results in altered gene dosage in both embryonic and extra-embryonic tissues reviewed in 30 ., In contrast , the WT/TG model reported here generates embryonic Dlk1 over-expression in the presence of a normal placenta with a normal Dlk1 dose , and the results indicate that Dlk1 can modulate embryonic growth independently of the placenta ., To complement these studies , it will be important to determine the extent to which placental Dlk1 expressed in the fetal endothelium and some trophoblast cells of the labyrinthine zone 10 also contributes to embryonic growth ., Importantly , this study shows that Dlk1 has a dual role on growth ., WT/TG neonates are born larger but fail to thrive during the first week of life ., The reduced postnatal growth rate is concurrent with reduced food intake in the neonatal period ., Maternally repressed imprinted genes are expected to favour growth performance and resource acquisition from embryonic stages to weaning according to the kinship theory which posits that imprinting arose as a consequence of a conflict between males and females over the allocation of maternal resources to the offspring 31 , 32 ., The embryonic overgrowth generated by Dlk1 double dosage follows the directionality predicted by this theory however the postnatal failure to thrive phenotype is contrary to it ., Our model implicates Dlk1 in reduced postnatal nutrient acquisition ., Indeed the results suggest that maternal repression of Dlk1 may have evolved to increase postnatal nutrient acquisition ., Furthermore it has been hypothesized that maternally repressed imprinted genes , such as Dlk1 , would minimize heating contribution within huddles 33 , 34 ., NST is not impaired in BAT over-expressing Dlk1 ., In fact , slight over-expression of Pparγ2 and Ucp1 may suggest the opposite scenario , whereby a maternally repressed gene may contribute more to the communal heating ., Our results are therefore inconsistent with the conflict hypothesis ., It is possible that imprinting may have evolved due to more than one type of selective pressure , perhaps different for different domains ., In TG/TG embryos over-growth occurs at E16 coincident with an increased frequency of embryonic mortality ., Embryos expressing this triple dose did not survive past birth ., A double dose of Dlk1 was compatible with embryonic viability but resulted in significant neonatal lethality ., We propose a negative correlation between gene dosage and survival that may fix an upper limit on growth promotion by Dlk1 ., One hypothesis for the lethality could be that Dlk1 dosage regulates the balance between proliferation and differentiation resulting in a trade-off between pre-natal size and developmental maturity ., Increasing Dlk1 dosage incrementally shifts the embryo towards increased growth , perhaps at the expense of organ maturation as was suggested for PatDi ( 12 ) embryos 16 ., The major abnormalities found in the liver , lungs and skeleton of the TG/TG fetuses may be signs of developmental immaturity of these organs ., The inability of the lungs to support TG/TG animals surviving to term , as well as pre-natal oedema , is consistent with this ., The Notch signalling pathway is required for branching morphogenesis and maturation of the lungs , and disruption of this pathway can lead to peri-natal lethality due to lung hypotrophy reviewed in 35 , 36 , 37 ., In contrast , except for subtle ossification delays in the WT/TG fetuses , no other clear signs of organ immaturity were identified , so we cannot conclude whether this contributes to the neonatal lethality of the hemizygous genotype ., In order to explore the cause of lethality in the WT/TG neonates more closely , we measured several parameters related to rodent well-being ., WT/TG animals were significantly less likely to have milk-filled stomachs on the day of birth ., This may be a result of defects in suckling behaviour engendered by many causes , such as appetite or olfactory regulation , or motor function ., Starvation may therefore be the cause of death of WT/TG neonates , and reduced feeding in the first week is also a likely cause of the reduced growth rate of surviving transgenic animals during this period ., Further analysis of the physiological consequences of this interesting growth trajectory of prenatal growth enhancement followed by postnatal compromised growth is in progress ., Imprinting of the Dlk1-Dio3 cluster is important ., Disruption of imprinted expression in several models results in lethality therefore there is a clear selective pressure to maintain imprinting at this cluster 13 , 38 ., The IG-DMR , like other imprinting control regions , controls the dosage of many linked genes ., It has been postulated that in imprinted clusters , some genes are the target of dosage regulation while other are merely bystanders whose altered dosage do not have phenotypic consequences and therefore are not targets for selection 39 ., We show that Dlk1 is a dose-dependent regulator of embryonic growth , as well as modulating processes necessary for postnatal survival ., WT/TG animals model loss of imprinting of Dlk1 , and have significantly reduced fitness , suggesting that the maintenance of imprinting of solely Dlk1 in eutherians by the IG-DMR has been sufficient to selectively retain imprinting control of the entire chromosome 12 cluster ., However , the phenotypes of PatDi ( 12 ) /PatDp ( dist12 ) and ΔIG-DMRMAT embryos are much more severe than that of a Dlk1 double-dose alone; therefore the dosage of at least one other gene in this cluster requires tight regulation ., Rtl1 is a likely candidate for this as defects have been reported in both knockout and overexpression models 38 and this gene is found only in eutherian mammals that have imprinting 12 ., We cannot know the selective pressures surrounding the acquisition of imprinting at chromosome 12 in eutherians however this mechanism must have evolved to regulate dosage-critical genes ., We have shown Dlk1 to be such a dosage-critical gene through its significant consequences on animal fitness upon dosage modulation ., It therefore could have been the target of selection for dosage control by imprinting at this locus ., Current levels of Dlk1 represent an optimal balance of maximized growth with minimal neonatal lethality ., TgDlk1-31 and TgDlk1-70 transgenes were obtained by restriction endonuclease mapping of the BAC clone 163O05 , screened from a BAC library of the mouse strain 129/Sv ., The two transgenes were microinjected individually into the male pronuclei of ( C57BL/6×CBA ) F1 zygotes and then implanted into foster ( C57BL/6×CBA ) F1 mothers ., The first generation of animals was obtained by crossing the founder with a ( C57BL/6×CBA ) F1 animal ., After the first generation , all transgenic animals were mated on a C57BL/6 background to maintain the lines ., Animals were housed four per cage ( maximum ) in a temperature-controlled room ( 24°C ) with a 12-hr light/dark cycle ., Food and water were available ad libitum ., All experiments involving mice were carried out in accordance with UK Government Home Office licensing procedures ., For the embryonic studies , the day of vaginal plug was considered day E1 ., RNA was extracted from whole embryos using TRI Reagent ( Ambion ) , following the manufacturers guidelines ., mRNA was then isolated from 120 µg of total RNA using Dynalbeads Oligo ( dT ) 25 kit ( Dynal ) following the supplied protocol ., We carried out northern-blot hybridization and used probes as described previously for Dlk1 and Gtl2 and Gapdh 3 ., For RT-qPCR , total RNA ( 10 µg ) was DNase-treated with RQ1 RNase-free DNase ( Promega ) following the manufacturers guidelines ., All cDNA was synthesized using random hexamers and Superscript III RNase H− Reverse Transcriptase ( Invitrogen ) , following standard procedures ., Taqman quantitative real-time PCR ( qRT-PCR ) was used to measure expression levels of Dlk1 and Gtl2 normalized to β-2-microglobulin ( β2m ) in different E16 and E18 embryonic tissues ., All RT-qPCR reactions were performed in a 25 µl final volume using standard Taqman qPCR conditions ( Appplied Biosystems protocols ) and amplified on a DNA engine Opticon 2 thermocycler ( MJ Research ) ., All reactions were conducted in triplicate ., Dlk1 expression levels were measured using the TaqMan gene expression assay ID - Mm00494477_m1 ( Applied Biosystems ) ., Gtl2 gene expression was measured using the forward primer 5′-GGGCGCCCACAGAAGAA-3′ , the reverse primer 5′- GGTGTGAGCCGATGATGTCA-3′ and the TaqMan MGB FAM probe FAM-5′-CTCTTACCTGGCTCTCT-3′-NFQ , spanning the Gtl2 exon 1-exon 2 boundary ., Finally , for the β2m gene , the forward primer B2M-32 5′-CACCCCCACTGAGACTGATACA-3′ , the reverse primer B2M-38 5′- TGGGCTCGGCCATACTG-3′ and the TaqMan MGB VIC-labelled fluorogenic probe VIC-5′-CCTGCAGAGTTAAGC-3′- NFQ were used ., Relative expression was calculated by normalisation to 100% using fetal E18 WT/WT tongue for the expression profile of the embryonic tissues and quantified using the comparative method ( 2−dCt ) 40 ., Pparγ2 and Ucp1 analysis was performed as previously described 41 ., The fluorescent signal emitted during PCR was detected using the DNA engine Opticon 2 sequence detection system ( MJ Research ) and post-PCR data analysis was performed using the Opticon Monitor analysis software version 2 . 02 ( MJ Research ) ., In situ hybridisation was conducted on paraformaldehyde fixed , wax embedded embryo and placenta sections at E16 and E18 of gestation according to the procedures previously described 42 ., Tissue lysates from whole embryos , whole placentas and multiple organs were used for SDS/PAGE analysis with a 10% polyacrylamide gel ., Resolved proteins were transferred to a poly-vinylidene difluoride ( PVDF ) Western blotting membranes Immobilon-P ( Millipore ) which were incubated with anti-DLK1 H-118 rabbit polyclonal antibody ( 1∶500 ) ( Santa Cruz Biotechnology ) or anti-DLK1 ( 1∶500 ) for placenta ( Proteintech ) and anti-α-TUBULIN mouse monoclonal antibody ( Sigma ) ( 1∶5000 ) overnight at 4°C , followed by the secondary HRP-conjugated antibodies: polyclonal goat anti-rabbit ( 1∶5000 ) or polyclonal goat anti-mouse ( 1∶7500 ) ( DakoCytomotion ) , respectively , on the next day ., Any signal was detected using ECL plus Western Detection System ( Amersham Biosciences ) ., The intensity of the signal was measured by scanning densiometry using Image J software ( NIH ) ., All embryonic dissections were recorded in terms of number and genotype of embryos/pups , dead embryos/pups , necrotic embryos and reabsorptions for all Dlk1

Introduction, Results, Discussion, Materials and Methods

Genomic imprinting is a normal process that causes genes to be expressed according to parental origin ., The selective advantage conferred by imprinting is not understood but is hypothesised to act on dosage-critical genes ., Here , we report a unique model in which the consequences of a single , double , and triple dosage of the imprinted Dlk1/Pref1 , normally repressed on the maternally inherited chromosome , can be assessed in the growing embryo ., BAC-transgenic mice were generated that over-express Dlk1 from endogenous regulators at all sites of embryonic activity ., Triple dosage causes lethality associated with major organ abnormalities ., Embryos expressing a double dose of Dlk1 , recapitulating loss of imprinting , are growth enhanced but fail to thrive in early life , despite the early growth advantage ., Thus , any benefit conferred by increased embryonic size is offset by postnatal lethality ., We propose a negative correlation between gene dosage and survival that fixes an upper limit on growth promotion by Dlk1 , and we hypothesize that trade-off between growth and lethality might have driven imprinting at this locus .

Genomic imprinting , the process that causes genes to be expressed from one of the two chromosome homologues according to parental origin , is likely to act on genes whose dosage is important for their correct function ., To test this , we compared the phenotype of transgenic mice expressing a double and triple dose of the imprinted gene Dlk1/Pref1 with animals expressing the normal single dose expressed from the paternally inherited chromosome ., Our results showed that a triple dose causes severe developmental abnormalities and death before or at birth ., Embryos expressing a double dose , recapitulating absence of imprinting , are bigger at birth but then around one-third of them died within the first three days of life ., Those that survived had poor early growth performance in the first week of life becoming small and remaining small , thus offsetting any benefit conferred by being born bigger ., Therefore , imprinted levels of Dlk1/Pref1 represent the optimal balance of growth versus lethality ., These findings lead to speculation about the evolutionary pressures acting to establish and maintain imprinting at this locus .

developmental biology/embryology, genetics and genomics/animal genetics, genetics and genomics/gene expression, developmental biology/developmental evolution, genetics and genomics/chromosome biology, developmental biology/molecular development, genetics and genomics/epigenetics

journal.pcbi.1000901

Similar Impact of CD8+ T Cell Responses on Early Virus Dynamics during SIV Infections of Rhesus Macaques and Sooty Mangabeys

The simian immunodeficiency virus ( SIV ) occurs as a natural infection in several Old-world monkey species , such as sooty mangabeys ( SM ) or African green monkeys 1 , 2 ., In striking contrast to HIV infection of humans , SIV infection does not cause disease in natural hosts ., The levels of virus replication , however , are similarly high in natural hosts and non-natural hosts such as rhesus macaques ( RM ) , in which SIV causes AIDS-like symptoms ., Comparative studies of SIV infection in natural and non-natural hosts provide the opportunity to investigate the interaction between the virus and the host immune system in pathogenic and non-pathogenic infection ., Such a comparison might shed light on the mechanisms of disease progression in pathogenic SIV and by extrapolation on HIV ., Although natural and non-natural hosts allow similar levels of virus replication , there are interesting immunological differences: SMs do not exhibit the increased CD4+ T cell turnover and the generalized immune activation that is characteristic for the SIV infection of RMs or HIV-infection in humans 3 , 4 ., Thus , virus load alone cannot be the key to understanding pathogenesis ., Silvestri and Feinberg 5 interpreted these findings in favor of the hypothesis that HIV disease progression is a result of generalized immune activation rather than of the destruction of CD4+ T cells by the virus alone ., This view of HIV pathogenesis is a derivative of the immuno-pathological hypothesis 6 ., Because primary HIV infection is a period critical for the future immune responses capability of controlling the infection 7 , 8 , the potential differences between pathogenic and non-pathogenic SIV infection are likely to manifest themselves early in infection ., In both RMs and SMs , the early SIV infection is divided into three phases ., The first phase is characterized by a sharp increase of virus load soon after infection ., The second phase describes the decline of virus load that follows the initial peak viremia ., The third phase finally describes the largely stable equilibrium virus load that eventually establishes after the decline ., This stable virus load is also referred to as the viral set point ., The characteristic pattern of virus load in primary SIV infection can be explained either through the delayed action of cellular immunity 9 , 10 or through target cell limitation 11 or both ., Note that in this context the term target-cell limitation refers to the hypothesis that the level of target cells on its own can explain the early virus-load dynamics 11 ., Regoes et al . 9 investigated these hypotheses by fitting mathematical models to viral loads of SIVmac239-infected RMs that exhibited either normal or experimentally impaired cellular immunity as a result of co-stimulatory blockade ., This analysis showed that target-cell limitation can explain the virus-load dynamics in the animals with impaired cellular immunity but not in those with a normal immune response ., In the latter case , the models could explain the virus-loads only if cellular immunity is also taken into account ., These results imply that target-cell limitation alone cannot explain the level of virus replication during primary SIVmac239 infection of RMs and thus suggest a role for cellular immunity in determining the post-peak decline of viremia ., In this article , we use the method of Regoes et al . 9 to analyze the early virus dynamics in non-pathogenic SIV infection of sooty mangabeys ( SM ) ., In particular , we sought to determine the roles that target-cell limitation , CD8+ T cell responses and NK cells play in primary infection of SMs , and to compare the impact of these factors with that in SIV-infected RMs ., To this end we fit the measurements of virus load with population-dynamic models that differ as to whether they take factors such as cellular immunity or NK cells into account ., Comparing the goodness of fit of these models , we can then evaluate the role of these factors in the primary infection of pathogenic and non-pathogenic SIV ., The analysis of the RM data reconfirms the results of Regoes et al . 9 in an extended dataset ., In particular , we find that target-cell limitation alone cannot explain the virus dynamics ., For all animals except one ( animal RPB8 ) , the best fit of the target-cell model predicts a steadily increasing virus load ( black lines in Figure 3 ) , i . e . the fit fails to explain the characteristic peak and the subsequent post-peak decline exhibited by the data ., Moreover , the quality of the fit is poor even for the animal for which the target-cell model can predict a viral load decrease ., Adding specific cellular immunity to the target-cell model does significantly improve the fit for RMs ( F-test , p\u200a=\u200a2 . 8×10−18 ) ., Importantly , the CD8+ T cell model can explain the characteristic post-peak decline of the viral load ( green lines in Figure 3 ) ., The results of our analysis of the data from SIV infection of SMs are strikingly similar to those obtained for the rhesus macaques: The target-cell model fails to explain the virus dynamics for all eight animals ( Figure 3 ) , whereas the CD8+ T cell model provides a significantly better fit ( F-test , p\u200a=\u200a1 . 3×10−11 ) , which can reproduce the qualitative patterns of the virus dynamics ., The only exception is the animal FSS , for which both the target-cell and the CD8+ T cell model produce poor fits ., The poor quality of these fits might be due to the fact that this animal exhibits a comparatively early increase of target-cell number and a comparatively late increase of CD8+ T-cell number ( see Figure 1 ) ., The similarity of the results in SMs and RMs suggests that the relative importance of specific cellular immunity and target-cell limitation during early infection is comparable in pathogenic and non-pathogenic SIV hosts ., In both cases , the temporal dependence of the viral load can only be explained if CD8+ T cells are taken into account ., Table 1 shows the best-fit estimates and the confidence intervals for the parameters of the CD8+ T cell model ., The parameters r and k quantify the per-cell impact of target-cells and CD8+ T-cells on the viral replication rate ( see equation 2 ) ., Both parameters are on average higher for sooty mangabeys: r roughly by a factor 6 and k by a factor 3 ., Furthermore , the intrinsic death rates of infected cells , a , were estimated to be 0 for most animals ., This suggests that , for both SMs and RMs , most deaths of infected cells are caused by cellular immunity ( see 9 ) ., The NK cell model and the CD8+ T cell & NK model are obtained from the target-cell and the CD8+ T cell model by adding NK cell number as an explanatory variable ., We consider the fits of these extended models for two reasons: First , to test whether the above results are robust against adding NK cells to the model and , second , to investigate the role of an important effector mechanism of the innate immune system during primary SIV infection ., In total , four types of statistical comparisons were performed ( see Figure 2 ) : Comparison, i ) between the target-cell model and the CD8+ T cell model is the one discussed above ., Comparison, ii ) between the target-cell model and the NK model evaluates whether adding NK cells to target-cell limitation improves significantly the quality of fit ., Comparison, iii ) between the NK model and the CD8+ T cell-NK model evaluates whether taking cellular immunity into account improves the fit of the NK model ., Finally , comparison, iv ) assesses whether NK-cell number does significantly improve the fit of the CD8+ T cell model ., NK cell counts were available for 8 SMs ( FWo , FYl , FWn , FFS , FRS , FSS , FUV , FWV ) and 4 RMs ( RPB8 , RSO8 , RYE8 , RZS8 ) ., If the number of all NK cells is used as a proxy of NK cell activity , extending the target cell based model by NK cells ( comparison, ii ) does improve the model fits significantly only for SM but not for RM ( F-test , p\u200a=\u200a0 . 016 and p\u200a=\u200a0 . 24 for SM and RM , respectively ) ., Extending the CD8+ T cell model by NK cells failed for both species to improve the model fits significantly ( F-test , p\u200a=\u200a0 . 98 and p\u200a=\u200a0 . 33 for SM and RM , respectively ) ., In contrast , extending the NK model by CD8+ T cells improves the fit significantly ( F-test , p\u200a=\u200a2 . 4×10−5 and p\u200a=\u200a5 . 7×10−4 for RM and SM , respectively ) ., If the number of proliferating NK cells is used as a proxy of NK cell activity , including NK cells again significantly improves the target-cell based model only for SM ( F-test , p\u200a=\u200a1 . 3×10−5 and p\u200a=\u200a0 . 33 for SM and RM , respectively ) ., In addition , including NK cell activity via this proxy also improves the CD8+ T cell model for SM ( F-test , p\u200a=\u200a0 . 00013 and p\u200a=\u200a0 . 97 for SM and RM , respectively ) ., These results suggest that NK cells play a role in the early infection of SM but not of RM ., The role of cellular immunity in early SIV/HIV infection has been a debated topic since the suggestion of Phillips 11 that early virus replication might be controlled by target-cell limitation ., Several lines of evidence suggest however that cellular immunity is an important force for the control of early SIV replication ., First , the post-peak decline of virus load coincides temporally with the rise of CTLs 12 ( although this is also consistent with the alternative explanation of 11 ) ., Second , 10 have shown that the post-peak decline of virus-load is significantly weakened if CD8+ T-cells are depleted ., Third , the ubiquitous selection for mutants that escape CTL response 13 also suggests an important role of cellular immunity ., Fourth , it has been shown that the patients ability to control HIV depends strongly on the alleles at the HLA and KIR loci 14 , which control the action of CD8 T cells and NK cells , respectively ., More recently , some of the authors of this paper 9 have shown that mathematical models can explain the early virus dynamics if they take both target-cells and CD8+ T-cells into account , but not if they take only target cells into account ., Our study extends this previous work by considering the impact of NK cells , important effectors of innate immunity ., In addition to the extended analysis of the early viral dynamics in pathogenic SIV infection , we here compare our results to non-pathogenic SIV infection in sooty mangabeys ( SMs ) ., This comparison has important implications for our understanding of pathogenesis ., Our analysis confirms the earlier finding of 9 that target-cell limitation alone cannot explain the virus dynamics in RMs ., We find that , in SIV-infected sooty mangabeys , target-cell limitation is equally unable to explain the viral load dynamics during early infection ., In both species , our model can only explain the virus dynamics if it takes cellular immunity into account ., This suggests that specific cellular immunity plays an important role in determining the dynamics of virus replication during early infection in both species ., We , however , also found that a model , which assumes a constant viral replication rate , independent of target cells , was unable to fit the virus-load data of all animals consistently ( results not shown ) ., This implies that , although target cells alone cannot explain the virus-load dynamics , in particular the peak and the post-peak decline , temporal variation of target cells is nevertheless important ., Overall , our results indicate that the relative impact of target-cell limitation and specific cellular immunity is similar in RMs and SMs ., These results give rise to testable predictions ., If , for example , one would selectively deplete NK cells during primary infection , the pattern of virus load should be affected in SM , but not in RMs ., In contrast , selective depletion of CD8+ T cells is predicted to lead to a loss of control of virus replication in both species ., Of note , all depletion experiments performed using an anti-CD8 antibody depleted CD8+ T cells as well as NK cells because both cell types express CD8 10 ., In RMs , treatment with a costimulatory inhibitor , which prevented the development of SIV-specific cellular and humoral immunity and reduced target cell levels , gave rise to target cell limited virus replication 9 ., The similarity between the factors governing virus replication predicts that an analogous treatment of SMs would also lead to target cell limitation ., Our conclusions about the role of cellular immunity and target-cell limitations are based on several assumptions ., First , the virus loads and the immune-cell densities were measured in the blood , which is not the main compartment of SIV replication and lymphocytes ., Our analysis , therefore , relies on the assumption that the measurements in the blood reflect the situation in the whole body ., In this context , it has been suggested that target-cell depletion in the gut might play an important role in the early SIV infection 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ., However , a recent study has shown that in SIV infections of both SM and RM , the target-cell depletion in the gut occurs too early to explain the peak in virus-load 23 ., Second , our models consider only the primary phase of SIV infection ., Therefore , our conclusion that cellular immunity does not differ in pathogenic and non-pathogenic SIV , does only apply to this phase ., It might thus be that cellular immunity at later phases plays a very different role in RMs and SMs , as suggested by numerous comparative studies 2 , 3 , 24 , 25 ., As discussed in Regoes et al . 9 , it is difficult to extend the approach used here to later phases of infection , because immune-escape and antibody responses would require considerably more complicated models ., Last , we cannot exclude that different cell compartments or cell types play the role of target cells in the SIV infections of sooty mangabeys and rhesus macaques ., Indeed , our model fits result in larger replication rate constants , r , for SM than for RM , which either suggests a better target cell utilization in SM , or is an indication that Ki67+ CD4+ T cells do not play the same roles in SM and RM ., Such an effect could systematically bias our analysis if our proxy ( i . e . proliferating CD4+ cells ) would be representative for target cells in one species but not in the other ., Finally , the p values of the model comparisons rely on the assumptions of normality and independence , which might be violated in our data ., Especially , autocorrelation in the virus-load and cell-numbers , might potentially lead to an overestimate of the degrees of freedom and thereby to an underestimate of those p-values ., However , it should be noted that independently of the statistical evaluation , the least-squares approach is a simple and intuitive method to fit dynamical models to data , and these fits clearly ( Figure 3 ) show that for all animals except one RM ( RPB8 ) , the best fit of the target-cell-limitation model fails to predict a post-peak decrease in virus-load ., This suggests that our results regarding the CTLs are robust against these ( in principle valid ) statistical concerns ., By contrast , adding NK cells to the model leads to smaller improvements of the fits and therefore these findings may be more vulnerable to potential autocorrelations ., One important caveat mentioned in the previous section is the uncertainty as to whether the measured cell populations ( e . g . Ki67+ CD4+ T-Cells , Ki67+ CD8+ T-Cells , NK cells ) can be identified with populations performing a specific function ( target cells , cytotoxic T cells , cytotoxic NK cells ) ., This potential problem is substantially alleviated by the way these measurements are integrated into our model ., Specifically , the quality of fit as measured by the residual sum of squares , is invariant with respect to a linear transformation of the variables ., I . e . if we measure the cell population x but the active population is x′\u200a=\u200aa x-b we will obtain the same quality of fit regardless of whether we incorporate x or x′ into our model ., Therefore it does not matter whether only a fraction of the measured cells is active or whether a constant number of the measured cells is inactive ., For practical reasons , however , it is important that the fraction of the active cells is not too small relative to the inactive cells , because then the noise in the latter is likely to overwhelm the signal in the former ., This reasoning implies that the comparison of the quality of fit of the different models ( Figure 2 ) is much more robust than the parameter estimates ( Table 1 ) : In principle , the first type of analysis ( model comparison ) still works , even if the linear transformation ( relating measured cell populations to the cell-populations performing a specific function ) is different for each animal ., By contrast , the second type of analysis ( parameter estimation ) requires that this transformation is similar in the animals compared ., For these reasons , we conclude that not too much weight should be given to the parameter estimates , as they rely much stronger on a good match between measured cell populations and the populations actually performing a certain function , while we can assert that the model comparison is robust ., The fundamental robustness of the method also explains why 9 found qualitatively similar results with Ki67+ CD8+ T-cells and tetramer positive T-cells as markers for SIV-specific cellular immunity ., As SIV infection is pathogenic in rhesus macaques but non-pathogenic in sooty mangabeys , our results can be interpreted in the context of current theories of SIV pathogenesis , in particular with respect to reasons underlying the absence of disease progression in SIV-infected SMs ., While an initial study suggested that acute SIV infection of SMs is characterized by limited to absent T cell activation 25 , a number of more recent studies that included a more comprehensive sample collection have shown very clearly that SMs exhibit substantial T cell activation during acute SIV infection 24 , 26 , 27 , 28 ., However , in marked contrast with SIV-infected RMs , sooty mangabeys are able to rapidly and dramatically reduce the level of T cell activation during the early chronic infection ( i . e . , starting at day 30 post inoculation ) 24 , 26 , 27 , 28 ., Although our model comparison did not directly test differences in the antigenicity of SIV between SM and RM , our results are more consistent with the latter observations and suggest that the divergent outcome of SIV infection in RMs and SMs is not caused by differences in CD8+ T-cell response during the early stages of infection ., All the experiments on non-human primates from which these data are sampled have been approved by the Institutional Animal Care and Use Committee ( IACUC ) ., All these experiments have been described in previous publications ., The data analyzed in this article were generated in experimental infections of SMs infected with the viral strain SIVsmm and of rhesus macaques infected with the strains SIVmac ( animals rbm , rvy , roz , ryt ) or SIVsmm ( animals RPB8 , RSO8 , RYE8 , RZS8 , RFT8 ) ., A detailed description of the experiments can be found in Garber et al . 29 , Gordon et al . 30 , and Mandl et al . 4 ., For the sake of comparability , we consider the same time-window as Regoes et al . , i . e . a window ranging from day 0 ( start of infection ) to day 30 ., In one of the rhesus macaques ( animal RFT8 ) no SIV infection could be established ., This animal was therefore excluded from further analysis ., In total , we consider 8 SMs ( all infected with SIVsmm ) and 8 RMs ( 4 infected with SIVmac239 and 4 infected with SIVsmm ) ., Figure 1 shows the measurements relevant for this study: the virus-load , the density of proliferating CD4+ T-cells , the density of proliferating CD8+ T-cells , and the density of NK cells ., The fraction of proliferating CD4+ and CD8+ T cells was assessed by staining for the nuclear antigen Ki67 , which is expressed by cycling cells ., We consider the density of proliferating CD4+ T-cells as representative for the size of the target cell population and the density of proliferating CD8+ T-cells as a surrogate measure for the SIV-specific cellular immunity ., We will therefore refer to the density of proliferating CD4+ T-cells and of proliferating CD8+ T-cells also as “target cells” and “cellular immunity” , according to the functional role we assume these populations to play ., Data on the density of NK cells were only available for all sooty mangabeys and for 4 out of 8 rhesus macaques ( RPB8 , RSO8 , RYE8 , RZS8 ) ., The data were analyzed by using population dynamic models , which describe the virus dynamics as a function of target cells , CD8+ T-cells , and NK cells ., The models are fitted to the virus load ., Hereby , the measurements of target cells , CD8+ T-cells , and NK cells were used as explanatory variables ., Importantly , the model does not aim to explain the measurements of these cell populations , but considers them only as factors that might explain viral replication ., A detailed account of this approach can be found in 9 ., In order to assess the role of target-cell limitation and cellular immunity in early SIV infection , we compared the fits of two nested models , which describe the virus dynamics by taking into account either target cells only or target cells and specific cellular immunity ., These models are referred to as the target-cell model and the CD8+ T cell model , respectively ., Mathematically , these models read ( 1 ) ( 2 ) where v is the virus load and T ( t ) and E ( t ) denote the number of proliferating CD4+ T-cells and of proliferating CD8+ T-cells , respectively ., The parameters r , a , and k are chosen for each animal such that T ( t ) and E ( t ) give the best possible prediction of v ( see below ) ., In order to test the impact of the non-adaptive immune system on our results , we extended the above models by adding NK cell number as an explaining factor ., We incorporate the impact of NK cells by using two different proxies: either the total density of NK cells ( characterized as CD3− CD20− CD16+ cells ) or only the density of activated NK cells ( i . e . Ki67+ NK cells ) ., The second approach is identical to the one used of CD8+ and CD4+ T-cells ., The first approach can be justified by the fact that , in contrast to CD8+ T-cells , NK cells do not recognize specific antigens ., Thus , every NK cell can potentially inhibit virus replication by either killing infected cells or by IFN-gamma production 31 , and their effect is most likely proportional to their level ., We would like to emphasize that we do not assume that every NK cell is cytotoxic , or that every NK cell has anti-viral activity ., We only assume that the impact of NK cells is proportional to their abundance ( see discussion ) ., The extensions of the target-cell model and the CD8+ T cell model are referred to as the NK-model and the CD8+ T cell & NK model ., Mathematically these models read ( 3 ) ( 4 ) where N ( t ) denotes the number of NK cells and the parameter n is chosen according to the best fit criterion ., We illustrate the fitting-procedure for the CD8+ T cell & NK model: First the differential equation of the model ( 4 ) can be integrated to ( 5 ) If t0…tk , denote the time points for which measurements of v are available then the parameters r , k , a and n are chosen such that the residual sum of squares ( 6 ) is minimized ., The integrals in the sum are computed from the measurements of the cell numbers T , E , and N by first interpolating these measurements by a piecewise linear function , resulting in the functions T ( t ) , E ( t ) , and N ( t ) , and then integrating these interpolating functions ., As expression ( 5 ) is linear in the parameters r , k , a and n , the best fit can be found using a standard linear-model solver such as the lm ( ) routine of the R language 32 ., Biologically , the parameters r , k , a and n must be larger than or equal to 0 ., If the best fit of ( 5 ) does not fulfill these conditions , one or several of the parameters r , k , a and n is set to 0 and the fitting procedure is repeated with these reduced functions ., From all the “reduced fits” , that one is chosen , which yields the minimal sum of squares while fulfilling the biological conditions ., The fits for the target-cell , the CD8+ T cell , and the NK model are obtained in a similar way as for the CTL-NK model ., In formula ( 5 ) the parameters that do not occur in the differential equation of the model ( i . e . equation 1 , 2 , or 3 for the target-cell , CD8+ T cell , and NK model respectively ) are set to 0 and the remaining parameters are chosen such that the corresponding sum of squares ( SSQtarget-cell , SSQCD8+ T cell , and SSQNK ) is minimized ., We can statistically compare two of the above models , for instance model 1 and model 2 , if they are nested , i . e . if model 1 results from model 2 by setting one of the parameters to 0 ., In such cases , model 2 will always provide a better fit than model 1 , because model 1 is included as a special case in model 2 ., Whether this improvement in the quality of fit is significant can then be assessed by performing an F-test ., The corresponding test statistic is Here SSQi denotes the residual sum of squares of the model i , and dfi refers to the corresponding degrees of freedom ., The p value that corresponds to the value of F is then calculated from the Fisher Distribution with degrees of freedom df1-df2 and df2 , i . e F ( df1-df2 , df2 ) ., This comparison between models can be made either for each animal individually , or , as we mostly do in this article , for all animals of a species taken together ., In the latter case , the residual sum of squares obtained by fitting the models to each animal and their corresponding degrees of freedom have to be summed to perform the F-test ., Figure 2 illustrates the statistical comparisons that are made in this article ., The most important of these comparisons is the one between the target-cell model and the CD8+ T cell model ( comparison i in Figure 2 ) , which assesses the relative importance of target cells and specific cellular immunity for explaining the virus-load dynamics ., If NK-cell counts are available , one can ask in addition whether taking NK cells into account improves the fit of the target-cell model ( comparison, ii ) , whether taking specific cellular immunity into account improves the fit of the NK model ( comparison, iii ) , and whether taking NK cells into account improves the fit of the CD8+ T cell model ( comparison, iv ) .

Introduction, Results, Discussion, Methods

Despite comparable levels of virus replication , simian immunodeficiency viruses ( SIV ) infection is non-pathogenic in natural hosts , such as sooty mangabeys ( SM ) , whereas it is pathogenic in non-natural hosts , such as rhesus macaques ( RM ) ., Comparative studies of pathogenic and non-pathogenic SIV infection can thus shed light on the role of specific factors in SIV pathogenesis ., Here , we determine the impact of target-cell limitation , CD8+ T cells , and Natural Killer ( NK ) cells on virus replication in the early SIV infection ., To this end , we fit previously published data of experimental SIV infections in SMs and RMs with mathematical models incorporating these factors and assess to what extent the inclusion of individual factors determines the quality of the fits ., We find that for both rhesus macaques and sooty mangabeys , target-cell limitation alone cannot explain the control of early virus replication , whereas including CD8+ T cells into the models significantly improves the fits ., By contrast , including NK cells does only significantly improve the fits in SMs ., These findings have important implications for our understanding of SIV pathogenesis as they suggest that the level of early CD8+ T cell responses is not the key difference between pathogenic and non-pathogenic SIV infection .

Simian immunodeficiency viruses ( SIV ) are typically non-pathogenic in their natural hosts ., However , if the same virus infects a non-natural host it often leads to AIDS-like symptoms ., Therefore , comparing SIV infections in these two types of host might help explain the pathogenesis of SIV in non-natural hosts and thereby also that of HIV ., We combined mathematical modeling with data on the levels of virus and immune cells early in infection , and compared both non-pathogenic SIV infections of sooty mangabeys and pathogenic SIV infection of rhesus macaques with respect to how the virus grows in them and to what extent it is controlled by the immune system ., We found that the impact of the immune system on early virus replication is remarkably similar in both species ., In particular , for both species virus replication can only be explained by the effect of CD8+ T cells ., These findings have important implications for our understanding of SIV pathogenesis as they suggest that the impact of the early immune responses is not the key difference between pathogenic and non-pathogenic SIV infection .

infectious diseases/hiv infection and aids, immunology/immune response, infectious diseases/viral infections

journal.pntd.0000590

A CATT Negative Result after Treatment for Human African Trypanosomiasis Is No Indication for Cure

Since none of the drugs for human African trypanosomiasis ( HAT ) is 100% efficacious , it is recommended to follow-up sleeping sickness patients every 6 months after treatment , for a period of 2 years ., Parasites may be difficult to detect in blood of HAT patients experiencing treatment failure , therefore assessment at follow-up visits relies mainly on lumbar puncture and examination of the cerebrospinal fluid ( CSF ) for presence of trypanosomes and white blood cell count ., A patient is declared cured when , within 2 years , no trypanosomes have been detected and the CSF white blood cell count returned to normal 1 ., Complete follow-up is seldom achieved because , when patients feel well , they are reluctant to comply to the follow-up examinations 2–5 ., So far , no markers for cure or treatment failure after HAT treatment have been identified in blood ., The card agglutination test for trypanosomiasis ( CATT ) is a fast and simple agglutination test for detection of trypanosome specific antibodies in blood of Trypanosoma brucei ( T . b . ) gambiense infected patients 6 ., With sensitivities between 87 and 98% and specificities of around 95% , the CATT test is extensively used in almost all HAT endemic areas for population screening , and has contributed to the current success of HAT control programs 7 , 8 ., Given the fact that drugs for HAT are toxic , and the specificity of CATT is limited , a confirmation step by parasitological techniques is needed 7 ., Trypanosome specific antibodies , detectable by CATT have been demonstrated even 24 months after successful treatment in no less than 47% of gambiense HAT patients 3 , 9 , 10 ., A positive post-treatment CATT result is therefore not indicative of treatment failure , but the predictive value of a negative CATT after treatment has hitherto not been evaluated ., We explored the hypothesis that a normalising , negative post-treatment CATT result indicates cure in gambiense HAT and rules out treatment failure ., If such CATT-normalising patients could be released from further follow-up , this would lead to major clinical and public health benefits as less lumbar punctures would be required and less patients should be followed for up to 24 months ., We report here on the pre- and post-treatment CATT serum results in a cohort of primary and retreatment HAT cases infected with T . b . gambiense ., Sleeping sickness patients originate from a prospective observational study ( THARSAT ) 11 ., The Commission for Medical Ethics of the Prince Leopold Institute of Tropical Medicine , Antwerp , Belgium and the Ethical Commission of the Ministry of Public Health , Democratic Republic of the Congo approved the study ., Written informed consent was given by all study participants prior to enrolment ., The cohort consisted of 242 primary HAT cases that had never been treated for HAT and of 118 retreatment cases previously treated for HAT , but with trypanosomes detected in CSF at inclusion ., All cases were parasitologically confirmed before enrolment and were ( re ) treated according to the national guidelines: primary cases in first stage ( n\u200a=\u200a41 ) were treated with pentamidine , primary cases in second stage were treated with melarsoprol ( n\u200a=\u200a192 ) or eflornithine ( n\u200a=\u200a9 ) ., Retreatment cases were treated with melarsoprol ( n\u200a=\u200a7 ) , eflornithine ( n\u200a=\u200a52 ) , melarsoprol nifurtimox combination therapy ( n\u200a=\u200a57 ) , melarsoprol eflornithine combination therapy ( n\u200a=\u200a1 ) or eflornithine nifurtimox combination therapy ( n\u200a=\u200a1 ) ., Patients were monitored for treatment outcome during 2 years ., The detailed description of the clinical outcomes in the cohort is given elsewhere 11 ., In brief , out of 242 primary cases , the final outcome was cure in 90 ( cure or probable cure ) and treatment failure in 118 ( relapse , probable relapse , or HAT related death during follow-up ) ., 34 primary cases were excluded from the analyses of post-treatment results since they could not be classified as cured or treatment failure because they were lost to follow-up , died during treatment or died over the following 2 years from non-HAT related causes ., Out of the 118 retreatment cases , 85 were cured and 16 experienced a new treatment failure ., Seventeen retreatment cases were lost to follow-up , died during treatment or died over the following 2 years from non-HAT related causes and were also excluded from the analyses of post-treatment results ., CATT/T . b . gambiense was performed following the titration-method as described by the manufacturers 6 on serum taken before treatment and at 3 , 6 , 12 , 18 and 24 months post-treatment ., The end titre ( highest dilution giving agglutination ) was determined ., Patients with end titres ≥1∶4 were considered CATT positive , end titres <1∶4 were considered CATT negative ., The Chi square test or Fisher exact test ( when the number of observations in a cell was <5 ) was performed for comparison of proportions using a 95% confidence limit ., Odds ratios ( OR ) with binomial 95% confidence intervals ( CI ) were computed ., STATA version 10 was used for data analysis ., The distribution of CATT end titres in primary and retreatment cases at inclusion is presented in figure, 1 . The median end titre in primary cases was 1∶16 ( interquartile range IQR 1∶8–1∶16 , mean±standard deviation: 14±9 ) , while it was 1∶4 ( IQR 1∶4–1∶8 , mean±standard deviation: 8±14 ) in the retreatment cases included in the cohort ., Sensitivity of CATT was 98 . 3% in primary cases ( 234/238 , CI 95 . 8–99 . 5% ) and 77 . 8% ( 91/117 , CI 69 . 2–84 . 9% ) in retreatment cases ., The median time between previous and current treatment in retreatment patients was 10 months ( IQR 6–16 months , data available for 103/118 cases ) ., The CATT results after treatment - in function of cure or treatment failure- are shown in figure, 2 . In the 90 cured primary HAT cases , the median end titre decreased to 1∶8 ( IQR 1∶4–1∶8 ) and 1∶4 ( IQR <1∶4–1∶8 ) after 3 and 6 months respectively , and became <1∶4 afterwards ., As shown in figure 2 , the proportion of CATT positives decreased in the cured group over time to 52% ( 45/87 ) and 37% ( 30/81 ) at 6 and 12 months and to 18% ( 15/83 ) at the final follow-up visit at 24 months ( also called test of cure ) ., In the 118 primary cases who experienced treatment failure within the 2 years of follow-up , the median end titre decreased to 1∶8 ( IQR 1∶4–1∶8 ) after 3 months and 1∶4 ( <1∶4–1∶8 ) after 6 and 12 months ., The proportion of CATT positives also decreased in function of time to 67% ( 44/66 ) and 62% ( 13/21 ) at respectively 6 and 12 months after treatment ., No significant relationship between CATT positivity and occurrence of treatment failure ( p>0 . 05 ) could be observed at 3 , 6 and 18 months post-treatment ., A significantly higher proportion of treatment failures cases tested positive with the CATT compared to the cases that were cured 12 months ( Chi square test , p\u200a=\u200a0 . 040 ) and 24 months ( Fisher exact test , p\u200a=\u200a0 . 027 ) after treatment ., The odds of a treatment failure case being CATT positive are 2 . 76 ( 95% CI 1 . 03–7 . 4 ) and 13 . 6 ( 95% CI 1 . 32–140 ) times greater than the odds of a cured case being CATT positive 12 and 24 months after treatment ., In 7/113 primary cases trypanosomes were detected in blood at time of relapse ., Two of them relapsed at 3 months with CATT titres 1∶8 and 1∶16 ; three others showed a titre 1∶4 ( relapses at 6 and 12 months ) and two relapsed at 24 months with titres 1∶8 and 1∶16 ., In the 85 retreatment cases who were cured after the current treatment , the median end titre was 1∶4 ( IQR <1∶4–1∶8 , IQR <1∶4–1∶4 ) after 3 and 6 months , and became <1∶4 afterwards ., The proportion of CATT positives decreased over time to 56% ( 45/81 ) and 36% ( 28/77 ) at 6 and 12 months and 17% at the 24 months test of cure visit ( figure 2 ) ., In 16 retreatment cases that experienced a repeated treatment failure after the current treatment , the median end titre was 1∶4 ( IQR <1∶4–1∶4 ) and the proportion of CATT positives 57% ( 8/14 ) 6 months post-treatment ., No significant relationship between CATT positivity and occurrence of treatment failure ( p>0 . 05 ) could be observed during follow-up ., In 3/16 of retreatment cases trypanosomes were detected in blood at time of relapse ., Their CATT titres were <1∶4 ( relapse at 3 months ) , and <1∶4 and 1∶16 ( relapses at 6 months ) ., We demonstrate for the first time that CATT sensitivity is low in retreatment cases , and that CATT titres decrease after treatment both in patients who experience treatment failure as well as in cured patients ., Before treatment , the CATT sensitivity in primary cases falls within the sensitivities previously reported for CATT in the Democratic Republic of the Congo , and for HAT in general 7 , 10 ., The low sensitivity of 78% observed in retreatment cases is explained by the decrease in CATT titre after a previous treatment , and largely corresponds to the proportion of CATT positives observed 6 and 12 months post-treatment within the groups of treatment failures ., The observed end titers are relatively low , being in respectively 39% and 83% of the primary and retreatment cases below 1∶16 ., Treating serological cases based on a CATT end titer ≥1∶16 , without parasitological evidence , might miss some HAT cases ., Although it has been shown that trypanosome specific antibody concentrations in blood of cured patients may persist up to 2 years or longer after treatment 3 , 9 , 10 , 12 , reports about the concentrations of specific antibodies in serum of sleeping sickness patients who experience treatment failure are rare ., In 22 relapsing patients , Frézil et al . 12 describe that the immunofluorescence test remains positive in the majority of relapsing cases , but doubtful/negative in only 1 case ., In a small cohort of 32 relapse cases , Miézan et al . 3 describe a decreasing antibody concentration and a CATT positivity rate of 94% at the moment relapse is diagnosed ., A negative CATT after unsuccessful treatment might be explained by trypanosomes that are cleared from peripheral tissues , such as lymph and blood , but that survive in the brain and thus do not trigger specific antibody production in the blood ., Our study has a number of limitations ., As a consequence of the diagnostic procedure used by the HAT control program to detect HAT , the observed sensitivity of CATT of 98% in our group of primary cases might be higher than in other patient cohorts ., Indeed , the patients in our cohort were identified as follows: CATT on whole blood , alongside cervical lymph node palpation , was used as a screening test and only those persons with a CATT positive result on whole blood , or having enlarged cervical lymph nodes underwent parasitological examinations for case confirmation ., Although part of the false negatives in CATT will be found by cervical lymph node palpation , the true sensitivity of CATT in the primary cases might be lower than 98% ., The number of treatment failures detected after ≥12 months is low , which prevents us from making reliable estimates of a further de- or increase in CATT titres after that time point , nor of the proportion of CATT positives ., Moreover , follow-up examinations in this cohort- as in routine clinical care- were focused on cerebrospinal fluid examination and the need for blood examinations may have been given less importance by the nursing staff ., As a consequence the cohort does not allow us to check if relapsing patients with trypanosomes in the blood had higher CATT titres than those without , since in the majority , relapse was confirmed by finding of trypanosomes in the CSF and no further blood examinations were performed ., Finally , the majority of primary cases in this study were treated with the trypanolytic drug melarsoprol in an area of high treatment failure rates ., Although a similar trend was observed in retreatment cases , who were treated differently , we cannot exclude that results could differ in primary HAT patients treated with other drugs ., Our findings have 2 practical implications ., First , the considerable proportion of CATT negative results in cases experiencing treatment failure , which increases over time , implies that a post-treatment CATT negative result does not necessarily indicate cure ., Knowing , moreover that a post-treatment CATT positive result does not indicate treatment failure , makes us conclude that CATT is unreliable for monitoring treatment outcome ., Secondly , screening programs for HAT should take into consideration that a careful history about past HAT episodes is paramount , as the sensitivity of CATT in relapse cases is not optimal ., Cases experiencing treatment failure are more likely to be false negative in CATT than new cases and , as a consequence , might be missed ( i . e . not offered parasitological investigations ) if they show no clinical signs ., These data might cast some doubt on the performance of CATT as a screening test in the detection process , given the fact that some relapse cases appear to be negative in the CATT ., Molecular -or other- diagnostics might eventually be taken up in an improved algorithm for diagnosis or follow-up but further investigation of these tests is necessary .

Introduction, Methods, Results, Discussion

Cure after treatment for human African trypanosomiasis ( HAT ) is assessed by examination of the cerebrospinal fluid every 6 months , for a total period of 2 years ., So far , no markers for cure or treatment failure have been identified in blood ., Trypanosome-specific antibodies are detectable in blood by the Card Agglutination Test for Trypanosomiasis ( CATT ) ., We studied the value of a normalising , negative post-treatment CATT result in treated Trypanosoma brucei ( T . b . ) gambiense sleeping sickness patients as a marker of cure ., The CATT/T . b . gambiense was performed on serum of a cohort of 360 T . b . gambiense patients , consisting of 242 primary and 118 retreatment cases ., The CATT results during 2 years of post-treatment follow-up were studied in function of cure or treatment failure ., At inclusion , sensitivity of CATT was 98% ( 234/238 ) in primary cases and only 78% ( 91/117 ) in retreatment cases ., After treatment , the CATT titre decreased both in cured patients and in patients experiencing treatment failure ., Though CATT is a good test to detect HAT in primary cases , a normalising or negative CATT result after treatment for HAT does not indicate cure , therefore CATT cannot be used to monitor treatment outcome .

The 2 year follow-up period required after treatment of human African trypanosomiasis ( HAT ) patients is a major challenge for patients and control programmes alike ., The patient should return every 6 months for lumbar puncture and cerebrospinal fluid examination since , so far , no markers for cure have been identified in blood ., The Card Agglutination Test for Trypanosomiasis ( CATT ) is a simple , rapid test for trypanosome-specific antibody detection in blood that is extensively used in endemic areas to screen for HAT ., We examined the value of a normalising CATT as a marker for treatment outcome ., We observed that CATT titres decreased after treatment both in patients who experienced treatment failure as well as in cured patients ., We conclude that CATT , though a good screening test , is unreliable for monitoring treatment outcome ., We also showed that the sensitivity of CATT in relapse cases was as low as 78% , and as a consequence some relapse cases might be missed in screening programs if they have no clinical signs yet .

infectious diseases/neglected tropical diseases, neurological disorders/infectious diseases of the nervous system

journal.pgen.0030053

A Tale of Two Oxidation States: Bacterial Colonization of Arsenic-Rich Environments

Although arsenic is most notorious as a poison threatening human health 1 , recent studies suggest that arsenic species may have been involved in the ancestral taming of energy and played a crucial role in early stages in the development of life on Earth 2 , 3 ., Further speculations involve this metalloid in the colonization of extraterrestrial environments containing high arsenic levels 4 , 5 ., Presently , arsenic contamination of drinking water constitutes an important public health problem in numerous countries throughout the world 6 ., Elevated concentrations typically derive from the weathering of arsenic-bearing minerals or from geothermal sources; lower amounts are of anthropogenic origin , e . g . , smelting and mining industries ., Microorganisms are known to influence arsenic geochemistry by their metabolism , i . e . , reduction , oxidation , and methylation 7 , 8 , affecting both the speciation and the toxicity of this element ., Arsenate ( AsV ) is less toxic than arsenite ( AsIII ) , but , paradoxically , resistance to AsV requires its reduction to AsIII , which will be extruded ., On the other hand , arsenite oxidation , which was primarily thought to constitute a detoxification mechanism 9 , may serve as an energy source in chemilithotrophic microorganisms 10 ., Bacteria metabolizing toxic elements represent therefore an attractive tool to restore contaminated sites ., In this respect , H . arsenicoxydans strain ULPAs1 , which oxidizes AsIII into its less toxic and more easily immobilized form AsV , has been proposed for use in the first steps of arsenic bioremediation 11 ., This heterotrophic microorganism , formerly called Caenibacter arsenoxydans ULPAs1 , was isolated from the activated sludge of an industrial water treatment plant contaminated with heavy metals such as arsenic , lead , copper , and silver 12 ., In the Burkholderiales order , its nearest phylogenetic relatives are members of the Oxalobacteraceae/Burkholderiaceae families , which contain several natural isolates with important biotechnological properties ., For example , bacteria of the Paucimonas 13 and Collimonas 14 genera are known for their polyhydroxybutyrate depolymerase and chitinase activity , respectively ., H . arsenicoxydans is a representative strain of a new genus comprising bacteria isolated from various aquatic environments , including contaminated , mineral and drinking water 12 , 15 , 16 ., To gain further insight into the mechanisms that permit the microbial colonization of arsenic-rich environments , we investigated the physiology of H . arsenicoxydans by genetic and functional approaches ., The results reported here , associated with descriptive and comparative genomics data , emphasize the metabolic versatility of this strain with regard to arsenic and the ability of microorganisms to restore liveable conditions within their ecological niche ., The H . arsenicoxydans genome consists of a single circular chromosome of 3 , 424 , 307 bp ( Figure, 1 ) with a total of 3 , 333 coding sequences ( CDSs ) , among which 38% are of unknown function ( Table 1 ) ., Surprisingly , any attempt to identify extrachromosomal elements in this strain by DNA sequencing , pulse-field electrophoresis , or plasmid purification was unsuccessful ( unpublished data ) ., This suggests that , unlike many microbes isolated from natural or anthropized environments , H . arsenicoxydans contains neither a second chromosome nor a ( mega ) plasmid ., In bacteria , mobile genetic elements are known to play a major role in the acquisition of genes involved in adaptation to environmental stresses 17 ., In line with this observation and in contrast to the situation in related microorganisms such as Ralstonia metallidurans 18 , only a small number of complete or partial insertion sequences ( IS ) were identified in the genome of H . arsenicoxydans ( Table S1 ) ., These IS elements represent 0 . 65% of the genome and belong to several families; i . e . , IS3 , IS30 , and IS110 ., H . arsenicoxydans also contains more complex transposons or transposon remnants , but , unlike those identified in biomining strains used for metal recovery from gold-bearing arsenopyrite ores 19 , they do not convincingly harbor arsenic-resistance determinants ., Interestingly , all the complete ISs are inserted with their transposase genes in a clockwise orientation with respect to the orientation of the replication fork , suggesting some sort of interference between replication and IS stability ., The overall GC content of the H . arsenicoxydans genome is 54 . 3 % but seven regions exhibit a lower content and one region a higher content than the average ( Figure 1 ) ., The presence of ISs was recorded in six of the low-GC modules ., Remarkably , the region with the highest GC content ( 63% ) , harbors several CDSs ( coding sequences ) identified as homologs of phage and/or plasmid-like genes coding for proteins involved in chromosome partitioning , DNA topoisomerase , DNA helicase , and DNA recombination and repair ( Table S2 ) ., This region , which covers ∼90 kb in the H . arsenicoxydans genome , is bordered by two tRNA genes at one end and one tRNA gene at the other end , and is flanked on one side by an integrase gene ( Figure 2 ) , suggesting a probable acquisition by an RNA-mediated horizontal gene transfer 20 ., In terms of similarity , this island is clearly made of three main parts ( Figure 2 ) ., The first one contains an arsenic-resistance cluster , also found in R . metallidurans , and , to a lesser extent , in Azoarcus sp ., and Pseudomonas fluorescens ., The second part of this region harbors a set of genes highly similar to part of the clc genomic island originally discovered in Pseudomonas sp ., strain B13 , which is known for its ability to degrade chloroaromatic compounds 21 ., The clc element is almost 100% identical over the whole length ( 102 kb ) to a chromosomal region in the chlorobiphenyl-degrading bacterium Burkolderia xenovorans LB400 22 ., Interestingly , a similar region of conserved synteny was observed in various proteobacteria such as R . metallidurans , P . fluorescens , Xanthomonas campestris , and Azoarcus sp ., , but none of the catabolic properties described in the clc element , mainly the clc and the amn operons ( allowing 3-chlorobenzoate and 2-aminophenol degradation , respectively ) , were found in this part of the H . arsenicoxydans island ( Table S2 ) ., Specific metabolic capabilities are found in the third part of this region , especially glutathione-dependent and -independent enzymatic activities involved in formaldelyde oxidation ., Finally , several resistance genes found in the clc element of the compared genomes ( e . g . , mercuric resistance in R . metallidurans or ultraviolet-light resistance in P . fluorescens , Table S2 ) support a role for this genomic island in the adaptive response to stressful environmental conditions ., Many bacterial strains of the Burkholderiales order are able to flourish in diverse ecological niches and grow on various carbon sources ., Surprisingly , H . arsenicoxydans can metabolize only a limited number of organic acids such as lactate , oxalate , succinate , and acetate; this is consistent with the presence of the corresponding functions on the chromosome and the absence of carbohydrate transporter genes such as those found in the phosphostranferase system ( Table S3 ) ., In addition , the use of amino acids as a sole carbon and nitrogen source is supported by the ability of the strain to grow on tryptone and the presence in its genome of multiple operons coding for amino acid transport systems ., In contrast , none of the pathways enabling carbon fixation from CO2 ( i . e . , genes coding for ribulose 1 , 5-biphosphate carboxylase/oxygenase and those involved in the Calvin cycle ) are present or complete , in conformity with the chemoheterotroph metabolism of H . arsenicoxydans ., Genes involved in the biosynthesis or degradation of glycogen were not identified in the genome of H . arsenicoxydans ., In contrast , the presence of the phbA-phbB gene cluster , which encodes a β-ketothiolase and an acetoacetyl-coenzyme A reductase , and phbC , coding for a poly-beta-hydroxybutyrate polymerase , is consistent with the accumulation in H . arsenicoxydans of poly-beta-hydroxybutyrate as an intracellular energy storage material ( unpublished data ) , as recently demonstrated in Ralstonia eutropha 23 ., Moreover , the genome contains all the genes encoding the inorganic phosphate transport and the phosphate-specific transport systems 24 , 25 , as well as genes possibly involved in the synthesis of high-energy polyphosphate granules ( Table S3 ) , which may constitute an additional means of energy storage for H . arsenicoxydans ., The diversity of electron transfer mechanisms is of prime importance in the management of energy in ecosystems subjected to frequent fluctuations in their oxygen content , such as water treatment plants ., Genomic data analyses suggest that H . arsenicoxydans can accommodate a wider range of oxygen concentrations than was initially anticipated 12 , 26 ., Indeed , the H . arsenicoxydans genome harbors multiple respiratory pathways , permitting microorganisms to grow under aerobic , microaerobic , and anoxic conditions ( Table S3 ) ., Reducing equivalents derived from organic compounds can enter energy-conserving electron transfer chains via a succinate dehydrogenase and three distinct formate dehydrogenases , none containing selenocysteine ., Possible inorganic electron donors are reduced sulfur compounds ( with the notable exceptions of sulphite and dimethyl sulphite ) and AsIII ., In contrast , no hydrogenase-encoding genes have been detected , suggesting that the strain may not gain energy from the oxidation of H2 to protons ., At the oxidizing end of bioenergetic electron transfer chains , five terminal oxidases might be operative ( Table S3 ) ., The two caa3 cytochrome oxidases usually operate under high oxygen tension while bo3 , cbb3 , and bd oxidases are more specific to low-oxygen conditions 27 ., All enzymes involved in anaerobic respiration via denitrification have been identified in the genome , i . e . , nitrate , nitrite , nitrous oxide , and nitric oxide reductases ., A β-proteobacterial cytochrome bc1-complex serves as a coupling site in many of these energy-conserving chains 28 ., The genome was explored to identify genes coding for cytochrome proteins possibly facilitating electron transfer between the Aox system and the bc1 complex and cbb3 cytochrome oxidase ., The consensus sequence for the cytochrome c center is Cys-x-x-Cys-His , in which the histidine residue is one of the two axial ligands of the heme iron ., Among the 56 putative heme-binding proteins we identified , Hear0476 was a particularly attractive candidate , because this protein was not predicted as a subunit of a cytochrome containing system , its coding gene was located immediately downstream of the aoxABC operon , and its expression was induced in the presence of arsenic ( Table 2 ) ., This putative protein belongs to the c552 family and was named AoxD ., Such a cytochrome has been shown to interact with the terminal cytochrome cbb3 in Helicobacter pylori 29 , to play the role of electron carrier to the bc1 complex in ammonia-oxidizing bacteria 30 , and to coprecipitate with AsIII oxidase protein in Alcaligenes faecalis 9 ., We therefore propose that AoxD represents the electron transfer link between AoxAB proteins and the ccb3 cytochrome oxidase and bc1 complex in H . arsenicoxydans ., Finally , any attempt to cultivate H . arsenicoxydans ULPAs1 with AsIII as an electron donor source was unsuccessful; this organism requires an organic compound as an energy source ., Moreover , neither selenate reductase nor respiratory arsenate reductase was identified in the genome ., In Shewanella sp ., , this latter enzyme is encoded by the arrAB locus and allows anaerobic respiration with AsV as a terminal electron acceptor 31 ., H . arsenicoxydans is not only resistant to arsenic but also to various heavy metals such as cadmium and zinc ( Table S4 ) ., This observation is consistent with the presence in its genome of multiple metal-efflux operons ( Figure S1 ) , e . g . , three cobalt-zinc-cadmium czc operons ., However , except for arsenic , the resistance levels to toxic metals were much lower than those measured in the metallophilic R . metallidurans ( Table S4 ) , which contains multiple plasmid-encoded genes 18 , suggesting a specific physiological adaptation of H . arsenicoxydans towards the arsenic ., Exposure to arsenic results in various biological effects , including DNA damage and oxidative stress 32 , 33 ., H . arsenicoxydans exhibits both positive oxidase and catalase activities 12 , in agreement with the presence of one catalase and two superoxide dismutase–encoding genes ( Table 2 ) ., The genome also encodes at least one thioredoxin peroxidase , one peroxiredoxin , one thioredoxin reductase , and one hydroperoxide reductase ., Moreover , genes coding for bacterioferritin and bacterioferritin comigratory protein , known to protect cells against toxic hydroxyl radicals resulting from iron overload , could also play a role in the adaptive response to oxidative stress 34 ., In addition , the partial screening of a Tn5-lacZ mutant library demonstrated an induction of several genes involved in DNA recombination and repair , e . g . , radA and polA , in the presence of arsenic ( Table 2 ) ., Their inactivation in H . arsenicoxydans led to an important loss of viability following UV exposure , which was further decreased by arsenic ( Figure S2 ) ., This suggests that the metalloid exerts a significant effect on DNA integrity ., Although arsenic methylation occurs widely in the environment , only a single bacterial methyltransferase ( ArsM ) , has been characterized thus far 35 ., Neither a homologous arsM gene nor arsenic methylation activity was detected in H . arsenicoxydans ( unpublished data ) ., In contrast , AsIII oxidation has been demonstrated in this organism 36 , resulting from the expression of the aoxAB operon ( Table 2 ) ., The AoxA-Rieske protein-encoding gene is located upstream from the AoxB catalytic subunit gene 36 ., Examination of available sequencing data , including those from the Sargasso Sea metagenome 37 , suggests a similar organization of putative AsIII oxidase genes in various microorganisms , e . g . , Thermus thermophilus , Chloroflexus aurantiacus , and Aeropyrum pernix ( Figure 3 ) , but not in the facultative autotrophic arsenite-oxidizing bacterium Alkalilimnicola ehrlichei MLHE-1 , in which no aox gene has been identified thus far 38 ., However , comparison of the neighboring CDSs revealed a limited synteny of other genes belonging to the aoxAB cluster in most organisms , even though an “arsenic genes island” as defined in Alcaligenes faecalis 10 ( i . e . , aoxAB genes close to arsenate resistance genes ars ) , was found in H . arsenicoxydans and in other organisms such as Nitrobacter hamburgensis and Chlorobium phaeobacteroides ., Remarkably , the second phosphate-specific transport locus identified in H . arsenicoxydans , which shows similarity with phosphate-specific transport systems in Serratia marcescens 39 and in Pseudomonas putida 40 , is located in the vicinity of the aoxRSAB locus ( Figure 3 ) ., This supports a link between arsenate , a structural analogue of phosphate , and phosphate transport ., Finally , homologs of aoxRS , a two-component signal-transduction system identified in Agrobacterium tumefaciens 41 , were found upstream from aoxAB in H . arsenicoxydans and most of the proteobacteria ( Figure 3 ) ., Their inactivation by transposon mutagenesis led to a complete loss of arsenite oxidase activity ( Figure S3 ) ., These genes were , however , not detected in the genome of other AsIII-oxidizing prokaryotes , which suggests important differences in aoxAB regulation among microorganisms ., Compared to most microorganisms , including R . metallidurans and the arsenic metabolizer Alkalilimnicola ehrlichei , which contain a single ars operon , H . arsenicoxydans is remarkable in that its genome harbors four different ars loci ( Figure 1 ) ., Indeed , three clusters of genes involved in resistance to arsenic were identified from a DNA genomic library in complementation experiments with an E . coli strain in which the ars operon was deleted ., They code for an ArsR regulator , an AsIII extrusion pump , an ArsH putative flavoprotein with no known function 42 , 43 , and one or two arsenate reductases ( ArsC ) , and confer a high level of resistance to arsenic ( Figure 4 ) ., Moreover , in silico analysis of the genomic data revealed the presence in H . arsenicoxydans of a fourth operon that lacks the AsIII pump-encoding gene and cannot therefore confer resistance to arsenate ., Its physiological raison dêtre remains enigmatic ., Quantitative analysis of the transporter-encoding gene mRNA demonstrated that the resistance operons are either constitutively expressed or induced in the presence of AsIII in H . arsenicoxydans ( Figure S4; Table 2 ) ., These observations were further supported by differential proteomic analyses of the ArsH , ArsC , and ArsR proteins preferentially accumulated on bidimensional gels in the presence of arsenic ( Figure S4;Table 2 ) ., Four arsenate reductases ( ArsCa ) in these operons belong to the group typified by the Staphylococcus aureus enzyme 44 ., The arsCa genes show a high sequence similarity and phylogenetic trees indicate that arsCa of loci 1 , 2 , and 4 arose from a recent gene duplication within the Herminiimonas lineage ( Figure 5 ) ., The same is true for the ArsR regulator and ArsH-encoding genes ( unpublished data ) ., Loci 2 and 3 operons additionally harbor a second reductase gene , arsCb ( Figure 4 ) , which is homologous to that of the E . coli R773 plasmid 45 ., An extensive analysis of available bacterial genome data shows that the simultaneous presence of arsCa and arsCb genes in arsenic resistance operons is common among α , β- , and γ-proteobacteria ., The frequent occurence of both arsCs in one operon argues against a mere redundancy of functions but is rather in favour of specific roles for each enzyme ., The S . aureus ArsC-type enzyme has been shown to use thioredoxin as a reductant 46 whereas the E . coli ArsC-type protein works with glutaredoxin 47 ., It therefore seems likely that the association of ArsCa and ArsCb enzymes enables various metabolic pathways to contribute reducing equivalents to the arsenic detoxification reaction , further enhancing the efficiency of this process ., Finally , the membrane transporter proteins differ strongly between the first two loci and locus 3 ( Figure 4 ) ., The latter contains an Acr3-type transporter 48 , which is typical for α- , β- , and several γ-proteobacteria ( Figure 5 ) ., The other loci both associate an ArsB-type transporter with ArsC reductases that cluster with homologous enzymes usually associated with the Acr3-type transporter ., Moreover , ArsB-type efflux pumps seem to be the main rule in Firmicutes , ɛ- , and γ- proteobacteria ( Figure 5 ) ., H . arsenicoxydans thus stands out among β-proteobacteria by its possession of two ArsB-type transporters in its ars operons ., Natural isolates have to constantly monitor the physicochemical parameters of their environment , which explains why numerous regulator-encoding genes have usually been identified in their genomes ., Except for histone-like nucleoid structuring protein ( H-NS ) , a RNA/DNA-associated protein widespread in proteobacteria 49 , the H . arsenicoxydans genome contains a complete set of genes coding for nucleoid-associated proteins such as HU , IHF , FIS , and Hfq ., No gene coding for catabolite activator protein was identified , consistent with the lack of carbohydrate metabolism in H . arsenicoxydans ., In contrast , the genome codes for the O2 responsive protein Fnr , enabling the modulation of the various respiratory pathways 50 ., Moreover , 42 genes coding for histidine kinases and response regulators were identified , which correspond to more than 1% of the whole genome ., They include those involved in the regulation of the resistance to or the detoxification of toxic metals and metalloids such as arsenic ( Figure S3 ) and presumably copper ( Figure S1 ) ., The genome also contains genes coding for a QseBC two-component regulatory system known to control flagellum synthesis and motility by quorum sensing ., No gene leading to autoinducer synthesis was identified , but this does not rule out the existence of an unidentified quorum-sensing system ., Flagellar genes are mainly clustered at two loci in the chromosome and were shown to encode a polar flagellum ( Figure S5 and Table 2 ) ., The rotation of this appendix is driven by sodium motive force , as demonstrated by the loss of motility of H . arsenicoxydans in the presence of 0 . 3 mM amiloride , an inhibitor of Na+/H+ antiporters ( Figure S6 ) ., Surprisingly , the synteny in the first locus is highly reminiscent of that in E . coli , whose motility is known to depend on peritrichous flagella 51 ., This region contains the flhDC master operon in H . arsenicoxydans , suggesting that the flagellum-encoding genes are organized in a mixed peritrichous/polar cascade in this organism ., Such a novel hierarchical cascade most probably results from gene acquisition from multiple sources followed by DNA rearrangements ., Although the flagellum morphology was not affected by the presence of arsenic ( Figure S5 ) , at least with respect to length and width , an increased concentration of this toxic element resulted in a concomitant increase in bacterial motility on semisolid agar plates ( Figure 6 ) and in flagellar gene expression ( Table 2 ) ., Aside from iron , an element that is essential to life , no such effect occurred with other toxic elements tested , such as CoII ( Figure 6 ) , CuII , SbIII , or AsV ( unpublished data ) ., The hypothesis that arsenic contributes to the metabolism of H . arsenicoxydans was further supported by the positive chemotactic response shown by the strain towards AsIII ( Figure 7A ) ., This observation suggests that the bacterium is able to sense and respond to the presence of AsIII in the medium ., The genome of H . arsenicoxydans contains 12 methyl-accepting chemotaxis proteins–encoding genes ., As most of these genes have no predicted function , it is tempting to speculate that at least one of them plays a role in this mechanism ., To determine how arsenic contributes to motility , GFP strains were constructed and their swimming behaviour was studied using video microscopy methods ., While the average swimming speed of H . arsenicoxydans was 30 μm/s , a 2-fold increase was observed in the presence of 2 mM AsIII ., Disruption of aoxA or aoxB gene by a transposon insertion located at the 84th or the 335th codon , respectively , abolished the improvement in the swimming performances in the presence of AsIII but not in the presence of FeIII ( Figure 6 ) , which suggests that the strain may gain additional energy from the arsenic-oxidation process ., The presence of aoxD , a cytochrome c552–encoding gene , in the vicinity of the aoxAB operon ( Figure, 3 ) further supports this hypothesis ., Finally , in contrast to related β-proteobacteria such as R . metallidurans , the genome of H . arsenicoxydans contains a type IV pilin gene cluster ., This mannose-sensitive haemagglutinin-encoding gene may be of importance in the interaction between H . arsenicoxydans and the microflora present in its environment ., Moreover , electron microscopy examination revealed the induction of a thick capsule when H . arsenicoxydans was cultivated in AsIII-containing medium ( Figure S5 ) ., An operon of 18 genes present in the genome , induced in response to arsenic ( Table, 2 ) and possibly involved in the synthesis of exopolysaccharides ( EPS ) , may play a role in this process ., In addition , nanoparticles were shown to accumulate in the capsule of H . arsenicoxydans as compared to H . fonticola , a phylogenetically related strain that does not oxidize arsenic ( Figure S7 ) ., Physicochemical analysis by transmission electron microscopy/energy dispersive X-Ray spectroscopy demonstrated the existence of a high arsenic content , suggesting for the first time a role for EPS in the scavenging of this toxic element ( Figure 7B ) ., Within the Oxalobacteraceae family , H . arsenicoxydans and the closely related strains H . aquatilis and H . fonticola represent a new genus comprising bacteria isolated from diverse aquatic environments ., Microorganisms of this novel taxonomic group may therefore be widespread in such natural or anthropized ecosystems ., The genome sequence and the physiology of H . arsenicoxydans further support the ability of this strain to grow in a wide range of environmental conditions , in particular with respect to oxygen concentrations ., Moreover , genomic and experimental data demonstrated that this organism is capable of accommodating the presence of high concentrations of various toxic metals ., More importantly , H . arsenicoxydans has evolved multiple processes not only to resist arsenic toxicity , such as DNA repair , oxidative stress resistance , and AsIII extrusion , but also to detoxify it for its own profit , such as AsIII oxidation and its probable involvement in energy metabolism ., Some of these unusual genetic determinants , acquired most probably by horizontal gene transfer , are organized as genomic islands ., Remarkably , the versatile regulatory system of H . arsenicoxydans enables it to sense dynamic changes in arsenic concentration and to initiate motility and EPS synthesis for attachment to this metalloid ., Such adaptive mechanisms may play a key role in the environment , allowing microorganims to efficiently flourish and colonize arsenic-rich ecosystems ., Moreover , recent results suggest that microbial biofilms are involved in the adsorption and immobilization of metals such as PbII 52 and CrIII 53 ., These properties have been used in biorehabilitation of aqueous solutions contaminated with heavy metals 54 ., The ability of H . arsenicoxydans to scavenge arsenic in an EPS matrix may be of prime importance in the context of bioremediation of contaminated environments , leading to the sequestration of this toxic metalloid ., To our knowledge , the genome of H . arsenicoxydans is the first to be fully characterized for an arsenic-metabolizing microorganism ., The results presented here provide evidence that the ability of microbes to colonize arsenic-rich environments extends beyond the biotransformation of this toxic element ., Although biochemical processes play an important role in arsenic release into the environment 7 , 8 , the physiology of the microbes inhabiting extreme ecological niches may not be restricted solely to oxidoreduction reactions ., In this respect , the positive chemotaxic response to the presence of arsenic and the scavenging of this element , associated with the transformation of AsIII into its less toxic form AsV , may have been key mechanisms in the colonization of the ancient environment on earth , allowing for the development of other microorganisms ., In the near future , sequencing data for other arsenic-metabolizing organisms , combined with molecular biology , genetics , biochemistry , and biophysics approaches , will lead us to identify new arsenic-dependent processes ., H . arsenicoxydans may therefore constitute a reference bacterium for further research towards a comprehensive analysis of the molecular mechanisms governing biological arsenic responses ., The complete genome sequence of H . arsenicoxydans was determined using the whole-genome shotgun method ., Three genomic libraries were constructed , i . e . , two plasmid libraries ( obtained after mechanical shearing of DNA and cloning of generated 3–4 kb and 8–10 kb fragments into plasmids pcDNA2 . 1 ( Invitrogen , http://www . invitrogen . com ) and pCNS ( a pSU18-derivative ) , respectively , and one BAC library to order contigs ( obtained by partial digestion with Sau3A of the genomic DNA and the introduction of ∼20-kb fragments into pBeloBac11 ( New England Biolabs , http://www . neb . com ) ., From these libraries , 26 , 112 , 7 , 680 and 3 , 840 clones , respectively , were end-sequenced using dye-terminator chemistry on ABI3730 sequencers ( Applied Biosytems , http://www . appliedbiosystems . com ) ., The Phred/Phrap/Consed software package ( http://www . phrap . com ) was used for sequence assembly and quality assessment 55–57 ., About 732 additional reactions were necessary to complete the genomic sequence ., Using the AMIGene software ( Annotation of MIcrobial Genes ) 58 , a total of 3 , 355 CDSs were predicted ( and assigned a unique identifier prefixed with “HEAR” ) and submitted to automatic functional annotation: BLAST searches against the UniProt database ( http://www . uniprot . org ) were performed to determine significant homology ., Based on the biological representation of the translation process , we applied Bayesian statistics to create a score function for predicting translation start sites ., We integrated together the ribosome binding site sequence , the distance between the translation start site and the ribosome binding site sequence , the base composition of the start codon , the nucleotide composition following start codons , and the expected distribution of proteins length ., To further increase the prediction accuracy , we took into account the predicted operon structures ., These elements were combined to create a score function and the highest score was selected for the translation start site predictions ( Y . Makita , unpublished data ) ., Protein motifs and domains were documented using the InterPro database ( http://www . ebi . ac . uk/interpro ) ., In parallel , genes coding for enzymes were classified using the PRIAM software 59 ., TMHMM v . 2 . 0 was used to identify transmembrane domains 60 , and SignalP 3 . 0 was used to predict signal peptide regions 61 ., Finally , tRNAs were identified using tRNAscan-SE 62 ., Sequence data for comparative analyses were obtained from the NCBI database ( RefSeq section , http://www . ncbi . nlm . nih . gov/RefSeq ) ., Putative orthologs and synteny groups ( i . e . , conservation of the chromosomal colocalisation between pairs of orthologous genes from different genomes ) were computed between H . arsenicoxydans and all the other complete genomes as previously described 63 ., Manual validation of the automatic annotation was performed using the MaGe ( Magnifying Genomes , http://www . genoscope . cns . fr ) interface , which allows graphic visualization of the H . arsenicoxydans annotations enhanced by a synchronized representation of synteny groups in other genomes ., The H . arsenicoxydans nucleotide sequence and annotation data have been deposited in the EMBL database ( http://www . ebi . ac . uk/embl; see accession numbers list below ) ., All these data ( i . e . , syntactic and functional annotations , and results of comparative analysis ) were stored in a relational database , called ArsenoScope 63 ., This database is publicly available via the MaGe interface at https://www . genoscope . cns . fr/agc/mage ., Mutations in genes induced by arsenic or mutant strains expressing GFP were obtained by random insertion of a mini-Tn5 and of a mini-Tn5 harboring a modified GFP-encoding gene 64 , respectively ., DNA manipulation and sequence analysis were performed as previously described 36 ., These analyses showed that the KDM-7 gfp mutant strain used in the present study carries a mini-Tn5 insertion in hear1692 , a CDS coding for a conserved hypothetical protein ., RNA preparation , probe construction , and quantitative analysis of transcripts were performed as previously described 65 ., The H . arsenicoxydans genomic library was constructed in plasmid pcDNA 2 . 1 ( Invit

Introduction, Results, Discussion, Materials and Methods, Supporting Information

Microbial biotransformations have a major impact on contamination by toxic elements , which threatens public health in developing and industrial countries ., Finding a means of preserving natural environments—including ground and surface waters—from arsenic constitutes a major challenge facing modern society ., Although this metalloid is ubiquitous on Earth , thus far no bacterium thriving in arsenic-contaminated environments has been fully characterized ., In-depth exploration of the genome of the β-proteobacterium Herminiimonas arsenicoxydans with regard to physiology , genetics , and proteomics , revealed that it possesses heretofore unsuspected mechanisms for coping with arsenic ., Aside from multiple biochemical processes such as arsenic oxidation , reduction , and efflux , H . arsenicoxydans also exhibits positive chemotaxis and motility towards arsenic and metalloid scavenging by exopolysaccharides ., These observations demonstrate the existence of a novel strategy to efficiently colonize arsenic-rich environments , which extends beyond oxidoreduction reactions ., Such a microbial mechanism of detoxification , which is possibly exploitable for bioremediation applications of contaminated sites , may have played a crucial role in the occupation of ancient ecological niches on earth .

Microorganisms play a crucial role in nutrient biogeochemical cycles ., Arsenic is found throughout the environment from both natural and anthropogenic sources ., Its inorganic forms are highly toxic and impair the physiology of most higher organisms ., Arsenic contamination of groundwater supplies is giving rise to increasingly severe human health problems in both developing and industrial countries ., In the present work , we investigated the metabolism of this metalloid in Herminiimonas arsenicoxydans , a representative organism of a novel bacterial genus widespread in aquatic environments ., Examination of the genome sequence and experimental evidence revealed that it is remarkably capable of coping with arsenic ., Our observations support the existence of multiple strategies allowing arsenic-metabolizing microbes to efficiently colonize toxic environments ., In particular , arsenic oxidation and scavenging may have played a crucial role in the development of early stages of life on Earth ., Such mechanisms may one day be exploited as part of a potential bioremediation strategy in toxic environments .

archaea, ecology, microbiology, computational biology, genetics and genomics, eubacteria

journal.pcbi.1000142

Evolutionarily Conserved Substrate Substructures for Automated Annotation of Enzyme Superfamilies

The molecular functions of enzymes result from a complex evolutionary interplay between environmental constraints , requirements for organismal fitness , and the functional malleability of a particular enzyme scaffold ., Within these constraints , existing enzymes are recruited during evolution to perform new or modified functions while often maintaining some aspects of the ancestral function 1–3 ., Consequently , among contemporary enzymes we observe groups of evolutionarily related enzymes that share some aspects of molecular function and differ in others ., The most divergent groups of evolutionarily related enzymes that still share aspects of function are called superfamilies ., Within a superfamily , we define a family as a set of proteins that perform the same overall catalytic reaction in the same way ., Why are some aspects of function shared and others allowed to change ?, By examining which aspects of function are shared among contemporary enzymes , we can gain insight into the requirements and constraints that govern this evolutionary process ., The focus of most studies of enzyme evolution has been the examination of conservation in sequence and structure ., The data available to conduct such studies is enormous and still increasing due to the multiplicity of ongoing genomic and metagenomic sequencing efforts 4 ., In tandem with the growth of sequence and structural data , a large number of new and sophisticated tools have been developed to improve our ability to identify the divergent members of superfamilies , allowing us to analyze patterns of conservation in sequence and structure that shed light on how enzyme functions have evolved and diversified ( for some examples , see 5–7 ) ., But such studies only capture aspects of enzyme evolution that can be inferred from the machinery that enables enzymatic catalysis , the enzymes themselves ., Far fewer studies have focused on the substrates and products of these reactions , with most of these focused on the requirements of metabolism 8 , 9 ., In this work , our goal is to understand the details of how enzymes function and evolve by studying the conservation and variation in their substrates and products ., In doing so , we aim for a more extensive view of enzyme evolution in order to improve our abilities to annotate enzymes of unknown function and to infer common aspects of function for superfamilies that have not yet been characterized ., The value of any analysis of the evolution of enzyme function depends on how we describe enzyme function , with respect to both the detailed molecular functions of individual enzymes and the properties of function shared across diverse members of enzyme superfamilies ., Previous approaches to study enzyme evolution range from detailed manual analyses of small numbers of related enzyme families and superfamilies to automated analyses of many superfamilies ., The former have often included not only analyses of sequences and structures but also comparisons of the substrates and reaction mechanisms of the constituent enzymes ., These studies have been useful for annotating new sequences and structures and for generating and testing hypotheses about patterns of enzyme evolution ( see 10–14 for examples ) ., However , because of the expert knowledge required and their time-intensive nature , these types of analyses are not feasible for large numbers of superfamilies ., Other semi-automated efforts have contributed to our understanding of enzyme evolution and data from these analyses have been made available in a number of online resources that include the Structure-Function Linkage Database 15 , MACiE 16 , the Catalytic Site Atlas 17 , and EzCatDB 18 ., Automated analyses more directly comparable to the large-scale and automated study described here 19–21 have used enzyme classification systems , like the Enzyme Commission ( EC ) system 22 , to represent functional properties and determine what properties are conserved ., The EC system represents a large proportion of known enzyme reactions , classifying each enzyme with a hierarchical set of four numbers that uniquely identify a reaction , and is easy to use for large-scale analyses ., However , this system , developed before analyses of enzyme evolution were common , does not provide a detailed description of enzyme function or substrates at the atomic level 23 ., Moreover , the EC classification of function often does not correspond with either the aspects of function that are conserved or those that can change during evolution ., These issues make this system unsuitable for evaluating how enzyme function evolves , especially when evolutionary relationships are distant 24 ., For enzymes , the Gene Ontology ( GO ) systems 25 molecular function classifications , also often used to describe and analyze function , largely recapitulate the EC system ., More similar to the work reported here , several groups have analyzed enzyme relationships and evolution using substrate and reaction similarities 26–28 ., Although these similarity metrics are useful , especially for clustering enzymes by their substrate similarities , they are not informative about what specific aspects of function are conserved , a specific goal of this work ., Here , we use graph isomorphism analyses to compare substrates of enzymes from 42 superfamilies to identify specific aspects of function conserved within each superfamily ., We also use comparisons of substrates and their corresponding products to determine whether and how much of the conserved substructure is involved in the reaction ., This comparison of substrates and products is similar to an analysis performed for a previous study with a different purpose , to predict EC numbers 29 ., To simplify the interpretation of results across the multiple superfamilies in this study , only enzymes comprised of single domains and that catalyze unimolecular reactions were investigated ., Automation of the analysis allows us to describe overall trends in functional conservation and variation across a large number of superfamilies ., A descriptive representation of conserved enzyme molecular functions using chemical structures and SMILES strings 30 , 31 is also provided ., This representation should be useful for annotating new members of superfamilies discovered in sequencing projects and for characterizing new superfamilies ., Results are presented for 42 superfamilies from the Structural Classification of Proteins ( SCOP ) database 32 ., These superfamilies meet the following criteria: ( 1 ) they consist of only single-domain enzymes that ( 2 ) perform only unimolecular reactions ( or reactions with two substrates , of which one is water ) , and ( 3 ) the superfamilies include at least two different reactions ( representing at least two different E . C . numbers ) for which substrate and product information are available in the enzyme database BRENDA 33 ., Sufficient data were available in BRENDA ( the third criterion ) for 46 . 2% of the superfamilies meeting the first two criteria ., These 42 superfamilies include representatives of six of the seven SCOP fold classes; the only fold class not represented is the membrane proteins class ., The enzymes in these 42 superfamilies represent a substantial proportion of the diversity of enzyme function , covering 25 . 4% of EC classes defined by the first two digits ( subclasses ) and 18 . 7% of EC classes defined by the first three digits ( sub-subclasses ) ., Conservation patterns were examined using only substrates and products as the data available in BRENDA were not sufficient to consider other aspects of reaction conservation , such as transition states and intermediates ., Our goal was to determine the molecular features that the substrates of a superfamily share and whether the shared features are involved in the reactions catalyzed by that superfamily ., Thus , for each superfamily , we identified the conserved substructure , defined as the set of bonds and their connected atoms that are present in all substrates of the superfamily ( Figure 1A ) ., These conserved substructures for the 42 superfamilies in our dataset are shown in Figure 2 ., Additional information about the diversity and conservation of functions in these superfamilies is provided in a hyperlinked table in the supplementary information online ( Table S1 ) ., Moreover , for each enzymes substrate ( s ) , we found the reacting substructure by determining what atoms and bonds change between the substrate and the product ( Figure 1B ) ., For each enzyme , we then determined whether the conserved substructure overlaps with the reacting substructure and by how much ., This overlap was quantified by calculating the fraction of the conserved substructure that is reacting ( fc ) ( Figure 1C , Table S2 ) and the fraction of the reacting substructure that is conserved ( fr ) ( Figure 1D , Table S2 ) ., Results for these measures of overlap are presented with respect to both the number of atoms and the number of bonds ., For a given superfamily , the average fc and fr calculated using atoms often differ from the values obtained using bonds ( Table S2 ) ., This difference arises because the number of bonds is frequently not proportional to the number of atoms in molecular structures ( e . g . , one bond consists of two atoms while three atoms can be connected by three bonds; a cyclic structure will have a different number of bonds compared to non-cyclic structure with the same number of atoms ) ., In addition , different types of reactions vary in the ratio of atoms and bonds that are involved in the reaction ( e . g . , a lyase may break one bond involving two atoms while an intramolecular transferase may involve one bond and three atoms ) ., Because both are valid measures of substructure size , both are provided in this report ., The distribution of average fc for the set of superfamilies ( Figure 3A ) indicates that there is a continuum among the superfamilies in how much of the conserved substructure is reacting , with superfamilies ranging from having little to having most of the conserved substructure participating in the reaction ., This trend is observed regardless of whether we use atoms or bonds in our calculations of average fc ., The results also show that all superfamilies with a conserved substructure have an average fc above zero , indicating that at least part of the conserved substructure is involved in the reaction ., Only one superfamily in our study set , the superfamily defined by SCOP as the metallo-dependent hydrolase superfamily , also known as the amidohydrolase superfamily 34 , 35 , has substrates so diverse that they do not share a common substructure of even a single conserved bond ., Detailed analysis of the superfamily , including analysis of differences in the overall functions , how active site motifs are used for catalysis , and other factors such as metal ion dependence , suggests that this group may be more properly considered as multiple superfamilies ( Brown and Babbitt , in preparation ) ., Plotting fr against fc illustrates the distribution of superfamilies ( Figure 3B ) across different patterns of overlap ( Figure 3C ) in the reacting and conserved substructures ., For simplicity , only the data calculated using atoms is provided in Figure 3B ., The values for each superfamily , calculated using both atoms and bonds , are provided in Table S2 ., The different regions in Figure 3B are intended merely to orient the reader to the range of variation across multiple superfamilies rather than to infer distinct categories implying fundamental differences between the superfamilies in different regions ., To determine whether there are differences in how a conserved substructure is used within a single superfamily , the variation of fc within each superfamily was also evaluated ( Table S2 ) ., Most superfamilies have little variation in how much of the conserved substructure is reacting ( Figure 4A ) ., However , there are a few superfamilies with substantial variation in fc ., We also evaluated the level of variation in which part of a superfamilys conserved substructure is used among the different reactions by calculating the average overlap in reacting and conserved substructures ( or ∩ c ) of every pair of substrates in the superfamily ., A flatter distribution and more variation was observed among the superfamilies for the average or ∩ c ( Figure 4B ) than for the standard deviation of fc ., The superfamilies that rank highest both in variation in fc and or ∩ c include the carbon-nitrogen hydrolase , metalloproteases ( “zincins” ) ( catalytic domain ) , and the thioesterase/thiol ester dehydrase-isomerase superfamilies ., Superfamilies that have low variation in fc and or ∩ c include the HD-domain/PDEase-like , dUTPase-like , and carbohydrate phosphatase superfamilies ., From these examples of superfamilies with high and low variation in fc and or ∩ c , we observe that the superfamilies with high variation tend to have smaller conserved substructures while superfamilies with low variation tend to have larger conserved substructures , though the correlation is not perfect ., The superfamilies in the low variation group have phosphate groups in the conserved substructure ., These tendencies may arise because different superfamilies and different types of reactions have different propensities for variation and conservation through evolution ., Alternatively , variation in how different superfamilies are defined in SCOP may lead to some of the variation observed among these superfamilies ., We also note that the set of reactions surveyed in this work represents only a subset of enzyme superfamilies , making it difficult to definitively address these hypotheses and questions ., More extensive analyses will be required to confirm and further explore these initial observations ., As new superfamily members are characterized , modifications of these substructure conservation patterns may be required ., To provide updates of this information , work is underway to incorporate this information into a searchable resource within our Structure-Function Linkage Database ( http://sfld . rbvi . ucsf . edu/ ) 15 ., Additional data generated in this study , including reacting substructures and how they overlap with conserved substructures for individual superfamily members , are available from the authors upon request ., As described below , our method can also be used to determine conserved functional characteristics for superfamilies that have not yet been characterized ., Programs and scripts required to perform these analyses are also available upon request ., By automating the analysis of enzyme substrates and reactions , the methodology introduced in this work facilitates the analysis of previously unstudied enzyme superfamilies ., This effort contrasts with previous analyses of enzyme superfamilies to determine patterns of functional conservation that have been highly labor-intensive , involving extensive manual analysis of reactions and literature-based curation of functional properties ( see the SFLD , http://sfld . rbvi . ucsf . edu/ , for examples ) ., The substructures conserved among the substrates of all members of a superfamily ( Figure, 2 ) provide annotation information that describes how function has been conserved in each of these superfamilies ., The certainty of these superfamily annotations will depend , however , on how well the range of substrates in each superfamily has been sampled ., Thorough substrate sampling may be especially critical for complex superfamilies that include many different catalytic functions ., While we have used all available reaction information in our analyses , the sampling of superfamily reactions may still be incomplete ., As new reactions are discovered through the sequencing of new genomes and metagenomes , these results can be updated and improved ., Despite these limitations , the characterization of superfamily-conserved substructures presented here facilitates the annotation of individual sequences on a large scale , helping to address the need for new strategies for automated function annotation ., This issue has become more pressing as the number of sequenced genomes increases and the era of metagenomics moves into high gear 41 ., Sequences that can be classified into a superfamily but not into a specific family can be annotated with the substructure common to all characterized members ., In these cases , often found in complex superfamilies exhibiting broad diversity in enzyme function , this may be the only level at which accurate annotation can be achieved , as insufficient information may be available to support annotation of a specific reaction or substrate specificity ., While substructure-based annotation does not by itself suggest a specific enzyme function , this information can be used as a starting point for additional analyses to determine specific function ., For example , many structures have been solved through structural genomics efforts , but their functions remain unknown 42 ., We have compiled a list of structures that have been classified into the SCOP superfamilies analyzed in this study , but have unknown functions ., These structures , many of them from structural genomics projects , can be at least minimally annotated with the substructure identified here as conserved across that superfamily , illustrated by the examples given in Figure 5 ( see Table S3 for the complete list ) ., Using this information , characteristics of ligands likely to be bound or turned over by these proteins can be inferred , providing guidance for biochemical studies to determine specificity ., These data also provide information about classes of small molecules that may be useful for co-crystallization trials to aid in solving the structures of these proteins or to capture them in functionally relevant conformations ., The variation found within superfamilies presents a caveat to be considered when using these substructures for function annotation ., While most of the superfamilies analyzed here have conserved substructures that are used consistently among the different superfamily members ( Figure 4 ) , there are a few superfamilies that have significant variation in the degree to which the conserved substructure is used in the reactions ., These superfamilies can be expected to be more difficult cases for function prediction since their variability makes it more difficult to determine conserved aspects of function ., In contrast , superfamilies with less variation in the degree to which the conserved substructure is used in the reaction are expected to be more straightforward cases for function prediction ., Understanding the patterns of functional conservation associated with the evolution of functionally diverse enzyme superfamilies can provide useful information for guiding enzyme engineering experiments in the laboratory 43 ., Using as a starting template for design or engineering an enzyme that already “knows” how to perform a critical partial reaction or how to bind a required substrate substructure ensures that some of the machinery required to perform a desired function is already in place ., Although still daunting , the task then simplifies to modifying the enzyme to bind and turn over a new substrate that contains the substructure consistent with the underlying capabilities of the superfamily ., As a corollary , aspects of function that have been conserved in all members of a divergent superfamily may be difficult to modify by in vitro engineering 43 , 44 ., Using such a strategy in a proof-of-concept study , two members of the enolase superfamily were successfully engineered to perform the reaction of a third superfamily member 45 ., As shown in Figure 6 , the superfamily-conserved substructure and the partial reaction associated with that substructure were not changed in these experiments ., Rather , engineering the template proteins to perform the target reaction involved changing each to accommodate binding the part of the substrate that is unique to the new reaction desired ., To allow for generalization of this approach , our analysis provides for all of the superfamilies that we investigated, 1 ) the parts of an enzymes substrate and reaction that are not conserved among related enzymes , which , provided they can be associated with regions of a target structure that interact with them , may point to structural features amenable to engineering , and, 2 ) the parts of the substrates that are conserved across all members of a superfamily , which may point to regions of the structure that may not be easily changed without loss of function or stability 46 ., In this study , requirements for a sufficiently large sample of enzyme reactions for a comprehensive analysis restricted us to using only substrates and products ., However , enzyme substrates can undergo intermediate changes during catalysis that are not adequately captured by looking only at substrates and products ., In some reactions , such as those in the enolase superfamily 47 , some portions of the substrate change and revert back to their original configuration during the reaction; these types of transformations are undetectable in the study described here ., The enolase superfamily represents a well-characterized example of chemistry-conserved evolution ., However , because our analysis does not currently detect such substrate changes , the average fc ( atoms ) for the enolase superfamily is 0 . 31 and the average fc ( bonds ) for the enolase superfamily is 0 . 34 , which places this superfamily in the middle of the distribution among our superfamilies for these measures of overlap ., Being able to detect the full extent to which structures change during a reaction would provide a better picture of substructure conservation in superfamilies like the enolase superfamily ., But this will require compilation of additional data to capture all of the partial reactions involved in a given overall reaction , including structures of reaction intermediates ., Emerging data resources , such as MACiE 16 and the SFLD 15 , currently seek to catalog information about reaction steps and mechanisms ., However , because this process is labor-intensive and often hampered by disagreement or ambiguity in the literature regarding the specific mechanisms of some reactions , these data resources are not yet sufficiently populated to support such broader analyses ., As these types of resources grow , we are optimistic that the information required to analyze reaction mechanisms more fully will become increasingly available ., Although it is beyond the scope of this study , correlating the conservation patterns we see in enzyme substrates with the conservation patterns in the sequence and structures of the enzymes themselves would also be a valuable extension for these analyses ., Finally , recent progress has been made in using in silico docking of small molecules to enzyme structures to infer molecular function ., In one such study , a library of high-energy reaction intermediates was generated and used to predict substrate specificity of enzymes in the amidohydrolase superfamily 48 ., As these methodologies are further developed , incorporation of predicted reaction intermediates into substructure analysis could improve prediction of substructures that are reacting ., In addition to benefiting from such recent advances in docking , the type of analysis presented here may in turn be used to improve applications of docking to predicting substrate specificity in enzymes ., Several such studies have recently focused on predicting functional specificity in the enolase 49 , 50 and amidohydrolase 51 superfamilies using knowledge about conserved substrate substructures from earlier analyses 15 , 52 to construct focused ligand libraries for docking ., We expect that the set of conserved substructures generated by our analysis can be used similarly to guide the construction of chemical libraries of ligands to improve prediction of substrate specificity in other superfamilies ., This study presents an automated method for analysis of superfamilies to determine the conserved aspects of their functions , represented by patterns of substrate conservation ., Our results show that superfamilies do not fall into discrete and easily separable categories describing how their functions may have evolved ., Rather , the conserved substructures determined in this analysis define superfamily-specific conservation patterns ., These results enable precise prediction of functional characteristics at the superfamily level for complex superfamilies whose members perform many different but related reactions , even when the evidence is insufficient to support more specific annotations of overall reaction and substrate specificity ., For applications in enzyme engineering , we expect that the identification of the aspects of function that have been most and least conserved during natural evolution will provide guidance for identifying the structural elements of a target scaffold that are most and least amenable to modification , thereby informing engineering strategies for improved success ., For our analyses , we used a subset of superfamilies from SCOP , a database of manually classified protein superfamilies , filtered based on criteria chosen to be most informative about enzyme evolution at high levels of functional divergence ., We included only superfamilies of single-domain enzymes with significant functional information in SCOPEC , a subset of SCOP with verified EC numbers , and in BRENDA , the most comprehensive database of enzyme experimental results ., Although many enzymes and proteins function as multi-domain units , the nature and organization of which can affect the specificity and regulation of enzymes 53 , for this study , we chose to use only single-domain enzymes as this allowed us to clearly assign a single function to one domain ., We included examples of enzymes known to have multiple structural domains only when the composite acts as a single functional unit ( e . g . , the enolase superfamily ) ., To ensure that the members of each superfamily were sufficiently divergent in function to analyze conservation of their substructures , only superfamilies annotated with at least two different EC numbers were investigated ., Compared to unimolecular reactions , bimolecular reactions have considerably more complex chemical and kinetic mechanisms for how substrates interact with the enzymes catalytic site ( i . e . , in what order different substrates bind ) ., Because these variations would have greatly complicated the analysis , we excluded superfamilies with any reactions that were not unimolecular ., Using the top level of the EC annotation , superfamilies were selected in which all the characterized members belong to any one of the following classes: hydrolases ( EC numbers 3 . x . x . x ) , lyases ( EC numbers 4 . x . x . x ) , and isomerases ( EC numbers 5 . x . x . x ) ., Experimentally verified substrate and product data were taken from the licensed version of the BRENDA database ( release 6 . 2 ) 54 ., Reactions were excluded in which ( 1 ) the product ( s ) had more than five ( non-hydrogen ) atoms more than the substrate or ( 2 ) substrates and products both had three or fewer ( non-hydrogen ) atoms ., Reactions in the first category are likely to be erroneous because they are not properly balanced ., Reactions in the second category are unlikely to be informative for the analysis because they contain so few atoms ., A “conserved substructure” ( Figure 1A ) contains the maximal sets of bonds in a substrate that are present in all the substrates of a superfamily , plus their adjacent atoms ., In all our analyses , we considered only bonds consisting of two atoms , neither of which is a hydrogen ., The “unconserved substructure” is the set of bonds in a substrate that are not in the conserved substructure , plus their adjacent atoms ., An atom can be in both the conserved and unconserved substructure if it is adjacent to both a bond in the conserved substructure and a bond in the unconserved substructure ., A “reacting substructure” ( Figure 1B ) consists of the bonds in a substrate that are not present in the product , their adjacent atoms , and any atoms that become connected in new bonds in the product ., In the case of a racemization reaction , in which the chirality of an atom center changes , the reacting substructure is defined as including the chiral atom that changes in the reaction , the four adjacent bonds and their adjacent atoms ., The “nonreacting substructure” is the set of bonds in a substrate that are also present in the product and their adjacent atoms ., An atom can be in both the reacting and nonreacting substructure if it is adjacent to both a bond in the reacting substructure and a bond in the nonreacting substructure ., The substrate substructure conserved among all characterized members of each superfamily was calculated using the maximal common substructure ( MCS ) algorithm implemented in the Chemistry Development Kit ( CDK ) 55 , an open source Java toolkit for manipulating small molecules ., The molecules are represented as graphs in which the nodes represent atoms and the edges represent bonds ., Each node is labeled with an atom type and each edge is labeled with the two atom types of the connected atoms and the bond order ., This algorithm finds , for a pair of molecules , the maximum common substructure ( MCS ) present in both molecules ., We extended this to find the MCS for the set of all known substrates for a superfamily ., In this initial analysis , we treated different atoms as dissimilar as long as the element type was different and bonds as different when the bond order and the two pairs of connected atoms were not identical ., The only exception to this rule was made for phosphate and sulfate groups , which we treated as similar in the substrate conservation analyses ., Our code allowed for the possibility of multiple unconnected MCSs by representing them as an unconnected graph with each connected portion corresponding to one MCS ., Although some of the pairwise MCSs contain multiple unconnected subgraphs , none of the superfamily-conserved substructures contain such multiple unconnected MCSs ., Finally , each substrate has a unique unconserved substructure defined as the set of edges not present in the conserved substructure and the atoms adjacent to these edges ., For each enzymatic reaction in which both the substrate and its corresponding product ( s ) are known , we calculated the non-reacting substructure by finding the MCS between the substrate and the product ( s ) ., The reacting substructure is the set of edges in the substrate that are not present in the product , plus the atoms adjacent to these edges ., The reacting substructure also includes atoms that form new bonds in the product ., To quantify the overlap between the reacting and conserved substructures , for each reaction in our dataset , we calculate fc ( Figure 1C ) , the fraction of the conserved substructure that is reacting and fr ( Figure 1D ) , the fraction of the reacting substructure that is conserved ., The values for fc and fr are calculated in two ways , using atoms or bonds , and the results for both are reported as they provide different but useful views of the data ., fc for bonds is determined by dividing the number of bonds that are in both the conserved and the reacting substructures ( r ∩ c ) by the number of bonds in only the conserved substructure ., fc for atoms is determined similarly , using the number of atoms instead of bonds ., Likewise , fr for bonds is determined by dividing the number of bonds that are in both the conserved and the reacting substructures by the number of bonds in only the reacting substructure; this value was also calculated using atoms ., For each enzyme in the BRENDA database , there may be multiple substrates with corresponding reactions that have been characterized ., For these cases , the values of fc and fr were obtained by averag

Introduction, Results, Discussion, Methods

The evolution of enzymes affects how well a species can adapt to new environmental conditions ., During enzyme evolution , certain aspects of molecular function are conserved while other aspects can vary ., Aspects of function that are more difficult to change or that need to be reused in multiple contexts are often conserved , while those that vary may indicate functions that are more easily changed or that are no longer required ., In analogy to the study of conservation patterns in enzyme sequences and structures , we have examined the patterns of conservation and variation in enzyme function by analyzing graph isomorphisms among enzyme substrates of a large number of enzyme superfamilies ., This systematic analysis of substrate substructures establishes the conservation patterns that typify individual superfamilies ., Specifically , we determined the chemical substructures that are conserved among all known substrates of a superfamily and the substructures that are reacting in these substrates and then examined the relationship between the two ., Across the 42 superfamilies that were analyzed , substantial variation was found in how much of the conserved substructure is reacting , suggesting that superfamilies may not be easily grouped into discrete and separable categories ., Instead , our results suggest that many superfamilies may need to be treated individually for analyses of evolution , function prediction , and guiding enzyme engineering strategies ., Annotating superfamilies with these conserved and reacting substructure patterns provides information that is orthogonal to information provided by studies of conservation in superfamily sequences and structures , thereby improving the precision with which we can predict the functions of enzymes of unknown function and direct studies in enzyme engineering ., Because the method is automated , it is suitable for large-scale characterization and comparison of fundamental functional capabilities of both characterized and uncharacterized enzyme superfamilies .

Enzymes are biological molecules essential for catalyzing the chemical reactions in living systems , allowing organisms to convert nutrients into usable forms and convert harmful or unneeded molecules into forms that can be reused or excreted ., During enzyme evolution , enzymes maintain the ability to perform some aspects of their function while other aspects change to accommodate changing environmental conditions ., In analogy to studies of enzyme evolution focused on conservation of sequence and structural motifs , we have examined a large number of enzyme superfamilies using a new computational analysis of patterns of substrate conservation ., The results provide a more nuanced picture of enzyme evolution than obtained either by detailed small-scale studies or by large-scale studies that have provided only general descriptions of function and substrate similarity ., The superfamilies in our set fall along the entire spectrum from the conserved substructure being mostly reacting to mostly nonreacting , with most superfamilies falling in the intermediate range ., This view of enzyme evolution suggests more complex patterns of functional divergence than those that have been proposed by previous theories of enzyme evolution ., The method has been automated to facilitate large-scale annotation of enzymes discovered in sequencing and structural genomics projects .

molecular biology/molecular evolution, genetics and genomics/bioinformatics, genetics and genomics/functional genomics

journal.pgen.1001287

Whole-Genome Comparison Reveals Novel Genetic Elements That Characterize the Genome of Industrial Strains of Saccharomyces cerevisiae

During its long history of association with human activity , the genomic makeup of the yeast S . cerevisiae is thought to have been shaped through the action of multiple independent rounds of wild yeast domestication combined with thousands of generations of artificial selection ., As the evolutionary constraints that were applied to the S . cerevisiae genome during these domestication events were ultimately dependent on the desired function of the yeast ( e . g baking , brewing , wine or bioethanol production ) , these multitude of selective schemes have produced large numbers of S . cerevisiae strains , with highly specialized phenotypes that suit specific applications 1 , 2 ., As a result , the study of industrial strains of S . cerevisiae provides an excellent model of how reproductive isolation and divergent selective pressures can shape the genomic content of a species ., Despite their diverse roles , industrial yeast strains all share the general ability to grow and function under the concerted influences of a multitude of environmental stressors , which include low pH , poor nutrient availability , high ethanol concentrations and fluctuating temperatures ., In comparison , non-industrial isolates such as laboratory strains , have been selected for rapid and consistent growth in nutrient rich laboratory media , thereby producing markedly different phenotypic outcomes when compared to their industrial relatives 3 ., The outcomes of these very different selection pressures are therefore most evident when comparing industrial and non-industrial yeasts ., As an example , laboratory strains of S . cerevisiae , such as S288c , are unable to grow in the low pH and high osmolarity of most grape juices and therefore cannot be used to make wine ., This is a clear difference between industrial and non-industrial strains of S . cerevisiae , however there are numerous subtle differences not only between industrial strains , but also between strains used within the same industry 4 , 5 , highlighting the overall genetic diversity found in this species ., There have been several attempts to characterize the genomes of industrial strains of S . cerevisiae which have uncovered differences that included single nucleotide polymorphisms ( SNPs ) , strain-specific ORFs and localized variations in genomic copy number 6–14 ., However , the type and scope of genomic variation documented by these studies were limited either by technology constraints ( e . g arrayCGH relying on the laboratory strain as a “reference” genome ) , or by the resources required for the production of high-quality genomic assemblies which has limited the scope and number of whole-genome sequences available for comparison ., In addition , to limit genomic complexity to a manageable level , previously published whole-genome sequencing studies on industrial strains used haploid representations of diploid , and often heterozygous , commercial and environmental strains 9–13 ., We sought to address these shortcomings by sequencing the genomes of four wine and two brewing strains of S . cerevisiae in their industrially-used forms ., The industries of winemaking and brewing were targeted for this work as they have the longest association with S . cerevisiae ( measured in the thousands of years ) and each industry has accumulated large numbers of phenotypically distinct strains for which genetic comparisons can be made ., This study demonstrates that industrial yeasts display significant genotypic heterogeneity both between strains , but also between alleles present within strains ( i . e . heterozygosity ) ., This variation was manifest as SNPs , small insertions and deletions , and as novel , strain and allele-specific ORFs , many of which had not been found previously in the S . cerevisiae genome and may provide the basis for novel phenotypic characteristics ., Interestingly , several ORFs were shown to comprise a gene cluster that was present in multiple copies and at a variety of genomic loci in a subset of the strains examined ., Furthermore , this cluster appears to have integrated into genomic locations by a novel circular intermediate , but without employing classical transposition or homologous recombination , which we believe represents the first time such an element has been characterized in S . cerevisiae ., Overall , this work suggests that , despite the scrutiny that has been directed at the yeast genome , there remains a significant reservoir of ORFs and novel modes of genetic transmission which may have significant phenotypic impact in this important model and industrial species ., Rather than being strictly diploid , many industrial yeast strains display chromosomal copy number variation ( CNV ) 18 ., In order to catalogue CNV in the industrial yeast genomes , the depth of sequencing coverage determined for each sequence contig were calculated such that areas of CNV could be detected as localized variations in that coverage ( Figure 1 ) ., There were several large areas of increased copy number across the strains including six potential whole-chromosome amplifications ( chrI of AWRI796 , chrVIII of VL3 , chrIII of FostersO and chrIII , V and XV of FostersB ) and one potential reduction in chromosomal copy number ( chrXIV of FostersO ) ., There were also several partial chromosomal CNVs , including amplification of 200 kb of chrXIV in AWRI796 , 600 kb of chrII and 200 kb of chrX in FostersO and a 400 kb reduction from chrVII of FostersO ( Figure 1 ) ., However , while the ale strains had a higher number of large CNVs than wine strains , the overall fold change of these CNVs was generally reduced ., This reduction can be most easily explained by the brewing strains having a polyploid genetic base while the wine strains are diploid , an observation which has been seen previously in these industrial yeasts 18 ., As existing published industrial yeast genome sequences were either generated from haploid derivatives of industrial strains 9–12 or had heterozygous regions discarded during analysis 13 , the level of genome-wide heterozygosity present in industrial strains remains largely unknown ., However , as the assemblies performed in this study retained genomic heterozygosity , it was possible to determine the level of allelic differences within each of these strains ( Table 2 ) ., While every industrial strain contained heterozygous single nucleotide polymorphisms ( SNPs ) , the proportion of these varied over thirty-fold between wine strain AWRI796 ( 1041 total heterozygous bp ) and the brewing strain FostersB ( 33071 bp ) ., Heterozygous insertions and deletions ( InDels ) were also present and ranged from single base pair variants to large InDels of up to 35 . 3 kb ., Strains were also shown to contain heterozygous instances of Ty element insertion , although , due to the repetitive nature of these elements , their presence in the genome could generally only be estimated through paired-end information ( data not shown ) ., In addition to the intra-strain variation that was present between homologous chromosomes within individual strains , there was also significant nucleotide variation between strains ., As seen for the allelic variation , both SNPs and InDels were found between strains , with inter-strain InDels of up to 45 kb being observed ., Many of the smaller InDels ( both heterozygous and homozygous ) were located in regions comprising tandem repeats ( Figure 2A , Table S1 ) and primarily in the expansion and contraction of di- and tri-nucleotide tandem repeats ( Figure 2B ) ., Indeed , when using chromosome XVI as an example , over 86% of the instances of di- and tri-nucleotide repeats displayed variable length in at least one of the strains ., As the size of tandem repeats has been associated with differences in gene expression 19 , this suggests that there are both strain and allele-specific differences in the expression of genes proximal to these repeat-associated InDel events ., SNP variation was also common throughout the strains with a total of 165 , 913 non-degenerate SNPs ( unique points of nucleotide variation ) that were present in at least one allele of the twelve strains investigated ( ∼1 . 3% of the total genome length ) ., However , given the influence of large , strain-specific InDels ( which were filtered out of the SNP analysis ) the apparent SNP density is much higher than 1 . 3% , such that these SNPs were shown to display a median inter-SNP distance of only 37 bp ., By using the number of SNPs separating any two isolates as an estimation of their relatedness ( Figure 3A ) , we were able to show that industrial yeasts are distinct from both the laboratory and human pathogenic strains and were also found to group by industry ., This was especially true of the brewing strains which displayed a high degree of genetic distance not only from the laboratory and human isolates , but also from the wine and bioethanol strains ., The only exception to this pattern of grouping by industry or environment niche was with the ‘natural’ isolate RM11-1a which grouped closely with wine strains ., However , given that it is descended from a strain sourced from a vineyard , RM11-1a may well share genetic origins with those strains used in winemaking ., In order to put the genetic variation observed in these genomic alignments in a larger population context , twelve strains were selected to represent each of the six main S . cerevisiae population groups as proposed by Liti et al 12 for further SNP comparison ( Figure 3B ) ., In this broader context , wine strains sequenced in this study were shown to also group tightly with the wine/European strains DBVPG1106 and DBVPG1373 , showing that the data produced across these two studies are directly comparable ., However , while the ale strains were still shown to be distinct from the wine isolates they were found to be far closer to the wine strains than isolates such as those used in sake production , which display the greatest level of nucleotide diversity when compared to the wine strains ., Indeed , when the SNP data from these additional strains in included in the calculations of SNP density , the total number of non-degenerate SNPs increases to 216 , 207 ( ∼1 . 7% ) with a median inter-SNP distance of only 27 bp ., However , despite comparisons to eighteen other diverse strains of S . cerevisiae 15 , 576 of these SNPs were found solely in this study ( 2 , 501 in more than one strain ) and with the vast majority of these SNPs being present in a heterozygous form ( only 1 , 864 novel SNPs were homozygous in at least one strain ) ., To determine how inter-specific variation at the nucleotide level translated into protein-coding differences , the predicted coding potential of each strain was compared ., ORFs were predicted from each sequence ( including the pre-existing whole genome sequences ) using Glimmer 20 and compared using a combination of BLAST 21 homology matches and genomic synteny to differentiate instances of orthology from gene duplication ( Table S2 ) ., When using the laboratory strain S288c as a reference , there was an average of 92% ORF coverage across the strains ., The majority of S288c ORFs without a match in other strains were shown to be located in repetitive regions of the S . cerevisiae genome such as in the sub-telomeric zones or the numerous Ty retrotransposons that are present in S288c genome relative to other strains ., Due to the repetitive nature of these regions it was often impossible to unambiguously position these sequences in the industrial yeast genome assemblies and they remain within repetitive , unmappable contigs in the various genome assemblies ., It therefore appears that , due to its persistent propagation in the laboratory , the genome of S288c may represent a reduced genomic state as it does not appear to contain additional genes that provide unique metabolic or cellular potential outside of those present in other strains ., It does however contain a far greater number of Ty transposons relative to all of the other strains suggesting that transposon proliferation occurred on at least one occasion during the development of this laboratory strain ., While the laboratory strain S288c is considered the reference for the genomic complement of S . cerevisiae , it is becoming apparent that it lacks a multitude of ORFs which exist in other strains of S . cerevisiae 9–13 , 22 , 23 ., This is confirmed n the present study with between 36 ( FostersB ) and 110 ( Lalvin QA23 ) ORFs lacking significant homology to the S288c genome but for which there were clear matches to sequences in other S . cerevisiae strains or microbial species ( Table S2 ) ., Orthologs of 102 out of 218 of the non-degenerate set of these ‘non-S288c’ ORFs have been identified previously in S . cerevisiae strains , mainly through whole-genome sequencing of AWRI1631 , EC1118 and RM11-1a and YJM789 8 , 9 , 13 ( Table S2 ) ., These include genes encoding proteins such as the Khr1 killer toxin 24 which is found in YJM789 , EC1118 , Vin13 , VL3 , FostersB and FostersO and orthologs of the MPR1 stress-resistance gene ( which was originally identified in the Sigma 1278b strain23 ) in RM11-1a , EC1118 , AWRI1631 , JAY291 , QA23 and VL3 ., Interestingly , in addition to these ORFs there were at least three proteins present in the human pathogen YJM789 and the FostersB and FostersO ale strains but which were lacking from the wine , biofuel and laboratory strains ( Figure 4C ) ., These included the YJM-GNAT GCN5-related N-acetyltransferase 8 and a separate gene cluster which is predicted to contain both RTM1 , which was identified previously as a distillery-strain specific gene that provides resistance to an inhibitory substance found in molasses 22 , and a large ORF of around 2 . 3 kb which , despite its large size and high-degree of conservation across the brewing and human pathogenic strains , lacks significant homology to any other protein sequences except for six isolates from the large S . cerevisiae population genomic screen which also appear to encode this protein 12 ( Figure S1 ) ., In addition to these two conserved ORFs , in the ale strains this cluster also appears to encode an invertase that would be expected convert sucrose into the sugars glucose and fructose ., Despite the presence of at least two existing high-coverage wine strain sequences and at least an additional six low coverage genomes , the entire repertoire of ORFs present in wine strains of S . cerevisiae , let alone the species as a whole , is far from complete ., In addition to expanding the strain range of previously identified non-S228c proteins , it was possible to identify at least eleven ORFs that lacked homology to existing proteins from S . cerevisiae , in addition to many new paralogs of existing S . cerevisiae genes ., These novel ORFs often clustered in large InDels , the largest of which was a 45 kb fragment in the wine strain AWRI796 ., This novel genomic region is located adjacent to a large repetitive element present on chromosomes XIII , XV and XVI , which hampered initial efforts to assign this region to a specific chromosome ., However , through the application of a 20 kb paired-end library , it was possible to bridge the repetitive region and position this novel region at the end of the right arm of chromosome XV ., This fragment is predicted to encode nineteen ORFs ( Figure 4A ) , three of which are predicted to encode aryl-alcohol dehydrogenases ( AADs ) ., AADs have been extensively characterized in filamentous fungi where they catalyze the reversible reduction of aldehydes and ketones to aromatic alcohols during lignin-degradation 25 , 26 ., These new AAD homologs are phylogenetically distinct from other AAD enzymes that have been identified , including the seven predicted AADs that are present in the S288c genome 27 , 28 ( Figure 4B ) ., One particularly curious feature of many of the industrial yeast strains analyzed in this study , was a cluster of five conserved ORFs that was present in all of the wine strains , RM11-1a and the bioethanol strain JAY291 , and potentially in at least four of the strains present in the Liti et al 12 study ( Figure 3 ) ., This cluster is predicted to encode two potential transcription factors ( one zinc-cluster , one C6 type ) , a cell surface flocullin , a nicotinic acid permease and a 5-oxo-L-prolinase , and has been suggested to be horizontally acquired by S . cerevisiae from Zygosacharomyces spp 13 ., In this study we have been able to show that while the sequences of the individual genes within this cluster are highly conserved between strains , the cluster itself is actually highly diverse with respect to copy number , genomic location and overall gene order ( Figure 5 , Table S3 ) ., The cluster was present in one to at least three copies across strains , with individual clusters being located in at least seven different genomic loci ( Figure 5A ) ., For example , wine strain Lalvin QA23 was shown to contain at least three copies of the cluster , found in three different genomic loci and with at least two copies being heterozygous ., However , despite this diversity , the sequence of the ORFs and intergenic regions of the cluster were highly conserved , with only fifteen nucleotide substitutions ( 0 . 01% ) recorded across the eleven known copies of the cluster ( Figure 5B , Figure S2 ) ., In addition to the differences in copy number and location , the exact order of the ORFs within the cluster differed in a location dependent manner ( Figure 5B , 5C ) ., However , all of these different ORF arrangements could be resolved into a syntenically-conserved order if the linear genomic copy of each cluster resulted from the differential resolution of a common circular intermediate , with a unique breakpoint in this circular arrangement being observed for each genomic location ( Figure 5B–5D ) ., However , despite the differential location of these clusters these integration events appear to select for functional conservation of the genes with the majority of the breakpoints being located within intergenic regions ( Figure 5B ) ., Of the two exceptions to this , one of these events occurs at the extreme 3′ end ( ∼100 bp from the predicted stop codon ) of one ORF such that a functional protein is likely to still be produced from this gene ., Adding further interest to the mode of transfer of this cluster , its integration into the genome appears to occur without the production of the terminal repeated sequences that would be expected if integration of this element occurred by either homologous recombination or classical mobilization via a transposon-like mechanism ., In fact , for at least three of the seven different integration events characterized in this study , integration of the cluster has occurred between two directly adjacent , conserved nucleotides , with a further two events showing only single nucleotide indels at the junction between the cluster and the flanking genomic sequences ( Figure 5E ) ., While S . cerevisiae is one of the most intensively studied biological model organisms and economically-important industrial microorganisms , many characteristics of its genome remain unknown , especially in strains other than the laboratory reference S288c ., Through the analysis of six industrial strains , it was possible to show that the industrial members of this species are distinct , with wine and brewing strains being almost as distantly related at the DNA level as they are to either the laboratory or human pathogenic strains ., This suggests that despite their roles in performing industrial fermentations , the two groups comprise genetically separate S . cerevisiae lineages ., While this is a situation similar to that proposed previously for wine and sake strains of S . cerevisiae 2 , the wine and ale strains were much more closely related to each other than to strains with origins outside of Europe 12 , and this may reflect a distant common European-type ancestor ., The bioethanol strain JAY291 displays an intermediate level of sequence relatedness to the wine strains ( compared to ale strains ) and also contains the five-gene cluster , suggesting that this strain shares at least some of its genomic origins with the wine isolates ., With the relatively recent development of the bioethanol industry , it is not entirely unexpected that yeasts used in this process may well have their origins in commercial strains used in established ethanologenic industries ., Wine strains would therefore make a logical choice for this starting point given their highly efficient production of ethanol and relatively high tolerance to a variety inhibitory substances , such as ethanol or polyphenols , that also exist in bioethanol fermentations 29 ., In addition to mapping the relationships between these strains , this study uncovered a number of genetic elements not previously identified in the S . cerevisiae genome , as well as expanding the range of several strain specific elements that had been identified previously ., This highlights the fact that the genetic variation that underlies the phenotypic diversity of S . cerevisiae goes well beyond that of SNPs or small InDels and is similar to the situation observed with many bacterial species where the pan ( species-wide ) genome is larger than that observed in any single strain 30 ., As for the situation observed with single nucleotide variation , several of these genetic elements link strains to specific industries ( e . g . the RTM1 cluster in the ale strains and the five-gene cluster in the wine strains ) ., It would therefore be expected that these ORFs provide selective advantage within specific industries that have favored their retention ., For some of these ORFs , such as the RTM1 cluster , the phenotypic benefits that they have historically provided in one industry may be advantageous in modern incarnations of others ., For example , modern wine production generally makes use of inoculated commercial strains ( rather than the historical use of wild yeast ) , which are produced on a large scale using molasses as a feedstock ., Genes such as the RTM1 cluster may therefore provide advantages in the production of modern commercial wine yeast , but which are lacking from the genomic complement of this group of strains due to the historical practices of winemaking ., While other strain-specific ORFs were shown to have much narrower strain ranges ( often single strains ) , it was possible to predict industrially-relevant roles for some of these genes ., For example , the novel AAD proteins that were identified in the wine strain AWRI796 may have a direct impact on the range of volatile aromas produced during fermentation , as the aromatic alcohols produced through the action of the AAD enzymes can present very different aromas profiles to their corresponding aldehydes and ketones 31 ., The presence of these AADs in specific industrial yeasts may therefore alter the profile of volatile aromas produced during winemaking or brewing , contributing to strain-specific aroma characteristics that are vitally important to many flavor and aroma-based industrial applications ., The role of ORFs such as those present in the wine yeast five-gene cluster are less clear but , given the potential regulatory role for at least two of these proteins , they could produce significant phenotypic effects ., The generally similar characteristics of high sugar and ethanol tolerance of Zygosacharomyces spp and the wine and bioethanol strains of S . cerevisiae 29 , 32 , may provide a selective advantage for growth under these conditions ., However , understanding the function of individual ORFs is overshadowed by questions regarding the origins of this novel cluster in addition to its effect on genome structure and dynamics ., It was recently proposed that this cluster entered the S . cerevisiae genome from Zygosacharomyces spp 13 ., Our data suggests that if this is the case , the transfer has either occurred on multiple occasions via a conserved circular intermediate that has integrated randomly into different genomic loci , or the fragment has entered the S . cerevisiae genome on a single occasion but has subsequently mobilized to new genomic locations via a circular intermediate ( Figure S3 ) ., Alternatively , this cluster is a mobile feature of the S . cerevisiae genome that has been lost from many strains and was transferred to Zygosacharomyces spp ., Regardless of the direction or precise mode of transfer it appears that this genetic cluster may mobilize throughout the genome via a method which has yet to be characterized in yeast and therefore provides an entirely new mechanism for the generation of variation in the S . cerevisiae genome ., A thorough understanding of the scope of plasticity of the yeast genome is a vital prerequisite for the systematic understanding of yeast biology or for the development of the next generation of yeasts for industrial applications ., As more S . cerevisiae strains are sequenced , the suitability of S288c as a “reference” strain for this species is becoming less clear , especially as it appears to lack a large numbers of ORFs found in many other S . cerevisiae strains while containing an abnormally high number of Ty transposable elements 8 , 9 ., Given the ubiquitous nature of the S288c genome for the design of ‘omics experiments , these novel elements have generally not been considered when studying strains other than S288c ., Thus , little data exists regarding the functional contributions of these proteins ., As such , they represent a significant knowledge gap with respect to cellular and metabolic modeling strategies ., This is especially true for proteins such as the ORF located next to RTM1 which is large ( ∼800 amino acids ) and highly conserved but has no significant homologs outside of a small subset of S . cerevisiae strains on which a function can be based ., Fortunately , the continued development of next generation sequencing , such as that applied in this work , have provided the means to now characterize large numbers of yeast strains to provide this information and outline the true scope and variability of this species ., Each commercial strain was obtained from the original mother cultures from the supplier ., Genomic DNA was prepared by zymolase digestion and standard phenol-chloroform extraction ., Library construction and sequencing was performed at 454 Life Sciences , A Roche Company ( Branford , CT ) using a pre-release development version of the GS FLX Titanium series shotgun and 3 kb paired-end protocols ., Sequences were assembled using MIRA ( http://sourceforge . net/apps/mediawiki/mira-assembler/index . php ? title=Main_Page ) and manually-edited using Seqman Pro ( DNAstar ) ., Regions of chromosomal CNV were determined by calculating the per-base sequencing coverage across each sequencing contig with median smoothing ( 1001 bp window , 100 bp step size ) ., The ratio between the coverage at each genomic location and the overall median genomic coverage was the calculated to determine the level of over-representation for each location ., Large-scale chromosomal aneuploidies were detected by screening for regions in which median ratio for a contiguous stretch of at least 101 individual segments differed from the overall genomic median by either 1 . 25 ( 5∶4 ratio representing at least 1 extra genomic copy in a tetraploid ) or 1 . 4 fold ( 3∶2 ratio representing at least 1 extra genomic copy in a diploid ) ., Chromosomal scaffolds from each yeast strain were aligned using FSA 33 ., Diploid sequences were assigned into two haploid alleles by converting any degenerate bases into their non-degenerate pairs ., Heterozygous regions were divided into both an insertion and deletion allele ., A chromosomal consensus was computed for the alignment based upon the most frequent allele at each position in the alignment ., Nucleotides that varied from the consensus in each strain were scored as sequence variants and were subsequently divided into SNPs ( nucleotide substitution ) or InDels ( nucleotide insertion or deletion ) ., To enable the comparison to strains with low coverage sequences 12 , SNPs that were calculated for each strain relative to S288c ( imputed SNPs ) were used to create synthetic S288c-based genome sequences that contain the SNPs present in these strains ., The genetic relationship between the strains was calculated by editing and concatenating the nucleotide alignments of all sixteen chromosomes using Seaview 34 followed by calculating the distance tree using the NJ algorithm of Clustalw ( ignoring gapped regions in the alignment ) ., Tandem repeats were predicted from the chromosomal alignment of all twelve yeast strains using Tandem Repeats Finder 35 using default parameters ( match weight , 2; mismatch , 7; indel , 7; pM , 0 . 80; pI , 0 . 10; minimum alignment score , 50; maximum period size , 500 ) ., Individual repeats were then scored as either being variable if the specific tandem repeat region contained strain- or allele- specific InDels ., ORFs were predicted using Glimmer 20 with the predicted ORFs of S288c being used to build the prediction model ( See Datasets S1 , S2 , S3 , S4 , S5 , S6 , S7 , S8 , S9 , S10 , S11 for actual CDS sequences for each strain ) ., Initial ORF designations were made by identifying the best sequence match for each ORF when compared to S288c using BLASTn 21 ., Glimmer was also used to predict ORFs from the sequence of S288c ( Accession numbers NC001133-NC001148 ) to correct for false-negatives in the predictions when compared to existing ORF designations in S288c ., ORFs with no match to S288c were searched against the full list of non-redundant Genbank proteins to identify a closest existing homology match ., ORFs from each strain were then arranged in syntenic order ( Table S2 for a full list of ordered ORFs ) ., For protein sequence comparisons , predicted protein sequences were aligned using Clustalw 36 ( http://align . genome . jp ) .

Introduction, Results, Discussion, Materials and Methods

Human intervention has subjected the yeast Saccharomyces cerevisiae to multiple rounds of independent domestication and thousands of generations of artificial selection ., As a result , this species comprises a genetically diverse collection of natural isolates as well as domesticated strains that are used in specific industrial applications ., However the scope of genetic diversity that was captured during the domesticated evolution of the industrial representatives of this important organism remains to be determined ., To begin to address this , we have produced whole-genome assemblies of six commercial strains of S . cerevisiae ( four wine and two brewing strains ) ., These represent the first genome assemblies produced from S . cerevisiae strains in their industrially-used forms and the first high-quality assemblies for S . cerevisiae strains used in brewing ., By comparing these sequences to six existing high-coverage S . cerevisiae genome assemblies , clear signatures were found that defined each industrial class of yeast ., This genetic variation was comprised of both single nucleotide polymorphisms and large-scale insertions and deletions , with the latter often being associated with ORF heterogeneity between strains ., This included the discovery of more than twenty probable genes that had not been identified previously in the S . cerevisiae genome ., Comparison of this large number of S . cerevisiae strains also enabled the characterization of a cluster of five ORFs that have integrated into the genomes of the wine and bioethanol strains on multiple occasions and at diverse genomic locations via what appears to involve the resolution of a circular DNA intermediate ., This work suggests that , despite the scrutiny that has been directed at the yeast genome , there remains a significant reservoir of ORFs and novel modes of genetic transmission that may have significant phenotypic impact in this important model and industrial species .

The yeast S . cerevisiae has been associated with human activity for thousands of years in industries such as baking , brewing , and winemaking ., During this time , humans have effectively domesticated this microorganism , with different industries selecting for specific desirable phenotypic traits ., This has resulted in the species S . cerevisiae comprising a genetically diverse collection of individual strains that are often suited to very specific roles ( e . g . wine strains produce wine but not beer and vice versa ) ., In order to understand the genetic differences that underpin these diverse industrial characteristics , we have sequenced the genomes of six industrial strains of S . cerevisiae that comprise four strains used in commercial wine production and two strains used in beer brewing ., By comparing these genome sequences to existing S . cerevisiae genome sequences from laboratory , pathogenic , bioethanol , and “natural” isolates , we were able to identify numerous genetic differences among these strains including the presence of novel open reading frames and genomic rearrangements , which may provide the basis for the phenotypic differences observed among these strains .

genetics and genomics/microbial evolution and genomics, genetics and genomics/comparative genomics

journal.pgen.1001152

H3K27me3 Profiling of the Endosperm Implies Exclusion of Polycomb Group Protein Targeting by DNA Methylation

Polycomb group ( PcG ) proteins are evolutionary conserved master regulators of cell identity and balance the decision between cell proliferation and cell differentiation 1 ., PcG proteins act in multimeric complexes that repress transcription of target genes; the best characterized complexes are the evolutionary conserved Polycomb Repressive Complex 2 ( PRC2 ) that catalyzes the trimethylation of histone H3 on lysine 27 ( H3K27me3 ) , and PRC1 , which binds to this mark and catalyzes ubiquitination of histone H2A at lysine 119 1 ., Plants contain multiple genes encoding homologs of PRC2 subunits that have different roles during vegetative and reproductive plant development 2 ., Whereas the EMBRYONIC FLOWER ( EMF ) and VERNALIZATION ( VRN ) complexes control vegetative plant development , reproductive development in Arabidopsis crucially depends on the presence of the FERTILIZATION INDEPENDENT SEED ( FIS ) PcG complex that is comprised of the subunits MEDEA ( MEA ) , FERTILIZATION INDEPENDENT SEED2 ( FIS2 ) , FERTILIZATION INDEPENDENT ENDOSPERM ( FIE ) and MSI1 2 ., The FIS PcG complex is required to suppress autonomous endosperm development; loss of FIS function initiates the fertilization-independent formation of seed-like structures containing diploid endosperm 3 ., In most angiosperms the endosperm is a triploid zygotic tissue that develops after fusion of the homodiploid central cell with a haploid sperm cell ., The endosperm regulates nutrient transfer to the developing embryo and regular endosperm development is essential for embryo development 4 ., Loss of FIS function also dramatically impacts on endosperm development after fertilization , causing endosperm overproliferation and cellularization failure , eventually leading to seed abortion 5 ., Thus far , only few direct target genes of the FIS PcG complex are known , among them the MADS-box transcription factor PHERES1 ( PHE1 ) 6 , FUSCA3 7 and MEA itself 8–10 ., All three genes are also targets of vegetatively active PcG complexes 7 , 11 , suggesting that different PcG complexes share at least a subset of target genes 7 ., Similar to extraembryonic tissues in mammals 12 , the endosperm has reduced levels of DNA methylation compared to the embryo or vegetative tissues 13 , 14 ., Hypomethylation is established by transcriptional repression of the maintenance DNA-methyltransferase MET1 during female gametogenesis 15 , together with active DNA demethylation by the DNA glycosylase DEMETER ( DME ) 13 , 16 ., Whereas the global DNA methylation levels differ only slightly between embryo and endosperm ( ∼6% for CG methylation ) , methylation differences at transposable elements and repeat sequences are significantly more pronounced 13 , 14 ., The functional significance of this genome-wide demethylation of the endosperm is not yet understood ., However , it has been proposed that DNA demethylation might cause transposon activation and generation of small interfering RNAs ( siRNA ) that might move to egg cell or embryo where siRNA-mediated DNA methylation would lead to increased methylation of parasitic genomic sequences 13 ., This notion is supported by the observation of accumulating 24nt siRNAs in the female gametophyte and in the endosperm 17 ., However , functional loss of RNA polymerase IV , the enzyme responsible for the biogenesis of siRNAs , does not cause reactivation of most transposons 18 , suggesting the presence of redundant pathways to silence transposable elements ., In this study , we profiled the H3K27me3 pattern in the endosperm and identified many target genes that were known previously to be targeted by vegetatively active PcG complexes , supporting the idea that different PcG complexes share a common set of target genes ., However , we also identified endosperm-specific H3K27me3 target genes that have functional roles in endosperm cellularization and chromatin architecture , suggesting that the FIS PcG complex has endosperm-specific functions and that PcG targeting in plants has tissue specific roles ., Finally and most importantly , we discovered that the FIS PcG complex in the endosperm targets transposable elements ( TEs ) that are protected by DNA methylation in vegetative tissues , implicating that DNA methylation and H3K27me3 are alternative repressive marks that may compensate for each other in the repression of a subset of TEs ., We established a transgenic line expressing PHE1 fused to the enhanced green fluorescent protein ( EGFP ) under control of the native promoter and 3′ regulatory elements ., Strong EGFP fluorescence was exclusively detected in endosperm nuclei from 1 day after pollination ( DAP ) until 4 DAP , whereas only a weak signal was detectable in the chalazal endosperm at 5 DAP ( Figure 1A ) ., EGFP-labeled nuclei from 1–4 DAP-old seeds were isolated with the use of a fluorescence-activated cell sorter ., High-throughput techniques allowed the harvesting , nuclei isolation , and sorting of approximately 100 000 nuclei in about 4 hours ., Within this time period , endosperm nuclei did apparently not undergo substantial changes in their transcriptional identity , as judged by a relatively low expression of embryo and seed coat marker genes in relation to the PHE1 gene ( Figure 1B ) ., Expression of seed coat and embryo marker genes followed a similar trend in microdissected endosperm samples ( Figure 1C ) ., To identify endosperm-specific PcG target genes we performed chromatin immunoprecipitation ( ChIP ) of chromatin from sorted endosperm nuclei using H3K27me3 specific antibodies followed by hybridization to high resolution whole-genome tiling microarrays ( Chip-on-chip ) ., As a control , we performed ChIP with unspecific IgG antibodies ., Genomic regions marked by H3K27me3 ( “H3K27me3 regions” ) were identified as continuous runs of probes with a MAT-score of at least 3 . 5 ( see Materials and Methods ) ., We identified 2282 regions that were significantly enriched for H3K27me3 , covering ∼1 . 9 Mb and representing ∼1 . 6% of the sequenced genome ., This corresponds to about one fourth the number of H3K27me3 regions identified in seedling tissues 11 , 19 , indicating that there are substantially fewer H3K27me3 targets in the endosperm than in vegetative tissues ., Similar to the H3K27me3 distribution in Arabidopsis seedlings 11 , most H3K27me3 regions in the endosperm were located on euchromatic chromosome arms and only 17 of the 2282 regions ( 0 . 7% ) were from centromeric or pericentromeric heterochromatin ( Figure 2A ) ., The distribution of H3K27me3 in endosperm over genes had a pronounced maximum in the transcribed region , similar to the distribution of H3K27me3 in vegetative tissues ( Figure 2B , 11 ) ., Notably , there was a small but distinct drop of H3K27me3 at the transcriptional start and shortly after the transcriptional stop , possibly caused by localized nucleosome depletion ., This interpretation would be in agreement with previous observations made in yeast and human cells , revealing nucleosome depletion at the transcriptional start and around polyadenylation sites 20–22 ., The length of H3K27me3 regions in the endosperm was comparable to the length of H3K27me3 regions in vegetative tissues 11 , with a median region size of about 750 bps ( Figure 2C ) ., MEA , PHE1 , MEIDOS ( MEO ) and FUSCA3 ( FUS3 ) as well as other genes that were previously identified as sporophytic H3K27me3 targets were among the endosperm H3K27me3 targets ( Figure 2D and Figure 3A ) , indicating that our procedure successfully identified H3K27me3 targets in the endosperm ., We identified 1773 genes to be associated with H3K27me3; of those , 1533 genes ( ∼86 . 5% ) overlapped with H3K27me3 marked loci identified in seedling tissues ( “shared H3K27me3 targets” ) 11 , 19 , whereas 240 loci ( ∼13 . 5% ) were specifically enriched only in the endosperm ( “endosperm-specific H3K27me3 targets” ) ( Figure 3A and Table S1 ) ., Most H3K27me3 targets in both sample sets are protein-coding genes of known or unknown functions , similar to the H3K27me3 targets in seedling tissues 11 , 19 ( Figure 3B ) ., The overall distribution of H3K27me3 marked pseudogenes and TEs in the endosperm and seedling tissues was similar; TEs and transposable element genes ( TEGs; correspond to genes encoded within a transposable element ) were clearly underrepresented among H3K27me3 targets compared to the genome average ( Figure 3B ) ., However , the frequency of TEs and TEGs was much higher among the endosperm-specific H3K27me3 targets than among the shared H3K27me3 targets , indicating that a subset of TEs and TEGs are specifically marked by H3K27me3 in the endosperm ( Figure 3B ) ., While 16% of all TEs and 46% of all TEGs probed by the microarray are located in centromeric and pericentromeric heterochromatin , only 5% of the TEs with H3K27me3 and 16% of the TEGs with H3K27me3 were from these heterochromatic regions ., Frequencies of almost all super families of TEs were similar among H3K27me3-marked endosperm-specific TEs and among all TEs detectable by the microarray ( Figure S1 ) ., Among the shared H3K27me3 targets LTR/COPIA ( p<5E-4 ) , LINE/L1 ( p<0 . 05 ) , and RathE1 elements ( p<0 . 05 ) were significantly enriched , indicating non-random targeting of TEs by PcG proteins ., We verified the specificity of our analysis by qPCR validation of endosperm-specific and shared H3K27me3 targets using independently prepared ChIP samples ., We randomly selected 10 endosperm-specific TEGs , 9 endosperm-specific genes and 8 shared target genes and could confirm all loci in an independent ChIP experiment ( Figure S2 ) , indicating that our procedure was specific with a low false discovery rate ., Shared H3K27me3 targets in the endosperm were highly enriched for genes involved in transcriptional regulation , with MADS-box transcription factors being a prominently enriched subclass of transcription factors ( p\u200a=\u200a3 . 01E-05; Table S2 ) ., However , many other GO categories were enriched among shared H3K27m3 target genes , including regulation of metabolism , flower development , cell wall organization , secondary metabolism and others ( Table S3 ) ., This indicates that the FIS PcG complex acts to repress a large set of genes that are not required during early endosperm development ., Among endosperm-specific H3K27me3 targets , there were many genes with potential roles in vesicle-mediated transport and cytoskeleton organization ( Table S4 ) , suggesting a specific function of the FIS PcG complex in endosperm cellularization ., Furthermore , many genes with functional roles in chromatin organization , such as the PcG protein encoding genes EMF2 , VRN2 , MSI1 , the DNA glycosylase ROS1 as well as DNA helicases were among specific H3K27me3 target genes ( Table S4 ) , implicating a role of the FIS PcG complex in establishing specific chromatin architectures in the endosperm ., Next , we analyzed the relation between H3K27me3 modification and gene expression ., Gene expression data were derived from the peripheral endosperm of seeds containing globular stage embryos , corresponding to the main fraction of the sorted endosperm nuclei used in our ChIP-chip experiment ., Consistent with the function of H3K27me3 in transcriptional silencing , the majority of shared endosperm H3K27me3 target genes were expressed at low levels ( Figure 4A ) ., In contrast , a fraction of the endosperm-specific H3K27me3 targets was moderately expressed ( Figure 4A ) ., Endosperm-specific H3K27me3 target genes had lower average H3K27me3 scores compared to shared targets independent of their expression level ( Figure 4B ) , suggesting that there is different efficiency of PcG protein targeting or PRC2 activity for endosperm-specific versus shared endosperm H3K27me3 targets ., Using publicly available datasets we tested the tissue-specific expression of endosperm-specific H3K27me3 target genes by cluster analysis ., Consistent with the idea that the FIS PcG complex is required for repression of target genes in the endosperm , genes present in clusters I , II and V ( 45% , n\u200a=\u200a75 ) were specifically repressed in the endosperm ( Figure 4C ) ., However , about half of all endosperm-specific H3K27me3 targets were expressed in the endosperm ( clusters III and IV , 55% , n\u200a=\u200a91; Figure 4C ) , in agreement with the higher average expression levels of endosperm-specific H3K27me3 target genes compared to non-H3K27me3 target genes ( Figure 4A ) ., We consider three not mutually exclusive explanations for this observation:, ( i ) H3K27me3 is not necessarily connected with gene silencing in the endosperm ., ( ii ) For a subset of genes only one of the alleles is marked by H3K27me3 ., In this case expression of the non-marked allele would be detected , whereas the H3K27me3 allele remains silenced , as it was shown before for PHE1 and MEA 8 , 9 , 23 , 24 ., However , imprinted genes predicted by Gehring and colleagues 14 were not among genes present in clusters III and IV ., ( iii ) PcG target genes are differentially regulated in the different domains of the endosperm , i . e . the micropylar , peripheral and chalazal domains ) ., TEs were strongly overrepresented among the endosperm-specific H3K27me3 targets compared to the shared H3K27me3 targets ( Figure 3B ) ., Hence , we hypothesized that the global DNA demethylation in the endosperm 13 , 14 caused H3K27me3 to accumulate in regions that are DNA methylated in vegetative tissues and , therefore , H3K27me3-poor ., This hypothesis predicts that TEs marked by H3K27me3 in the endosperm have reduced endosperm DNA methylation levels compared to all TEs ., Indeed , median endosperm CG and CHG DNA methylation levels were lower at H3K27me3 marked TEs than at other TEs ( Figure 5A ) ., CHH methylation levels were generally low and did not differ between H3K27me3 marked TEs and all TEs ( data not shown ) ., TEs that carried H3K27me3 in endosperm and vegetative tissues were almost devoid of CG DNA methylation in endosperm and vegetative tissues ., In contrast , TEs that carried H3K27me3 only in the endosperm had high DNA methylation levels in vegetative tissues while DNA methylation levels in the endosperm were markedly below the average over all TEs ., Similarly , shared TEGs were almost devoid of DNA methylation in vegetative tissues and in the endosperm ., Endosperm DNA methylation levels of specific H3K27me3 TEGs were comparable to the average DNA methylation levels in the endosperm of all TEGs present in the genome ( Figure 5B ) , indicating that reduced DNA methylation levels in the endosperm might allow targeting of PcG proteins to defined sequences independent of residual DNA methylation ., CHG methylation followed a similar trend as CG methylation ( Figure 5B ) ., In contrast , no substantial changes in CHH methylation levels were observed ( data not shown ) ., Protein coding genes were generally much less DNA methylated than TEs or TEGs ., Similar to shared TEs and TEGs , shared H3K27me3 target genes were almost devoid of DNA methylation in vegetative tissues and the endosperm ( Figure 5C ) ., In marked contrast , endosperm-specific H3K27me3 target genes had significantly higher CG DNA methylation levels in vegetative tissues than the genome-wide average ( Figure 5C ) , supporting the idea that CG DNA methylation prevents these genes being targeted by PcG proteins in vegetative tissues ., CG DNA methylation level of endosperm-specific H3K27me3 genes was reduced in the endosperm compared to vegetative tissues , again suggesting that reduced DNA methylation levels in the endosperm enable targeting of PcG proteins to selected loci ., Shared and specific protein coding H3K27me3 target genes were almost devoid of CHG and CHH methylation in vegetative tissues and the endosperm ( Figure 5C and data not shown ) ., Together , we conclude that DNA methylation and H3K27me3 , which both can bring about transcriptional repression of target genes , usually exclude each other at target chromatin ., In the endosperm , where DNA methylation is naturally reduced , some loci that were DNA methylated in other tissues become targeted by the FIS PcG complex and marked by H3K27me3 ., This hypothesis predicts that experimental reduction of DNA methylation levels in vegetative tissues will cause PcG proteins to be targeted to some loci that are usually DNA methylated ., Indeed , in met1 mutants H3K27me3 was found at some TEs that did not carry H3K27me3 in wild type 25 , strongly supporting this idea ., Based on their expression in the endosperm , two main clusters of protein coding genes and TEGs that were DNA methylated in vegetative tissues and carried H3K27me3 in the endosperm were apparent ( Figure 5D ) ; the first cluster contained genes and TEGs that were weakly expressed in other tissues and became specifically repressed in the endosperm , whereas the second cluster contained genes and TEGs that were mainly repressed in other tissues and became specifically expressed in the endosperm , indicating that loss of DNA methylation fostered expression of several genes and transposons in the endosperm independent of their gain of H3K27me3 ., We wondered whether loss of FIS activity would cause a global deregulation of H3K27me3 target genes ., Therefore , we profiled the fis2 transcriptome of seeds harvested at 3 DAP and 6 DAP and searched for deregulated genes that were marked by H3K27me3 in the endosperm ., Loss of FIS function profiled at 3 DAP and 6 DAP resulted in different and largely non-overlapping gene expression profiles ( Figure 6A ) ., Although the overlap of H3K27me3 target genes and genes deregulated upon loss of FIS function was significant ( p\u200a=\u200a3 . 0E-05 and 5 . 7E-04 for 3 DAP and 6 DAP , respectively ) , expression of surprisingly few target genes ( ∼1 . 5% and ∼1 . 8% at 3 DAP and 6 DAP , respectively ) was increased upon loss of FIS function ( Figure 6A , Table S5 ) ., EMF2 and VRN2 expression was not increased in fis2 seeds at 3 or 6 DAP , indicating that loss of FIS2 function is not compensated by increased expression of FIS2 homologous genes ., Genes deregulated at 3 DAP and 6 DAP fell into two largely distinct clusters ., Whereas most of early deregulated genes were not expressed in the wild-type endosperm until heart stage , late deregulated genes were predominantly expressed during early wild-type endosperm development and became repressed around heart stage ( Figure 6B ) , supporting the idea that the FIS PcG complex is required for the repression of a defined set of genes around endosperm cellularization 26 , 27 ., Genes deregulated in fis2 at 3 DAP and 6 DAP were prominently enriched for glycosyl hydrolases ( Table S6 ) , with a strong enrichment of Family 17 of plant glycoside hydrolases at 6 DAP ., Family 17 members preferentially hydrolyse the major component of endosperm cell walls , callose , 28 , suggesting that repression of cell wall degrading enzymes is a requirement for successful endosperm cellularization ., Conversely , this implicates that increased expression of these genes in fis mutants might contribute to the failure of fis mutant endosperm to undergo endosperm cellularization 29 ., Importantly , we did not detect increased expression of TEGs in fis2 mutants , suggesting that loss of H3K27me3 might be compensated by other repressive mechanisms ., If so , we wondered whether in seeds lacking both , FIS activity and CG DNA methylation , repression of TEGs would be relieved ., Therefore , we generated fis2/FIS2; met1/MET1 double mutants that contain 12 . 5% seeds homozygous for met1 and devoid of FIS activity ., We randomly selected eight endosperm-specific H3K27me3 TEGs ( At4g16870 , At5g37880 , At3g32110 , At2g13890 , At5g35710 , At1g35480 , At3g28400 , At2g16010 ) that were DNA methylated in vegetative tissues and had decreased DNA methylation levels in the endosperm ( Figure S3 ) ., Among those , At4g16870 , At5g37880 had increased expression levels in fis2;met1 double mutants compared to met1 and fis2 single mutants ( Figure 6C ) , whereas expression of At3g32110 equally increased in met1 and fis2; met1 double mutants ., Expression of the other TEGs was not significantly changed compared to wild type ( data not shown ) ., Based on these data we conclude that DNA methylation and FIS-mediated H3K27me3 can act synergistically to repress a subset of TEGs in the endosperm , but that there are additional mechanisms to silence TEGs in the absence of both mechanisms ., PcG proteins are largely viewed as general suppressors of genomic programmes that are not required in a specific tissue type or during a particular developmental stage of an organism 1 ., This would predict that a large set of PcG target genes is shared in different tissues , as only a small set of genes is expressed in a tissue-specific fashion 30 ., In line with this view , we found that the majority of PcG target genes identified in the endosperm are also targeted by PcG proteins in vegetative tissues 11 , 19 , suggesting that different PcG complexes share a common set of target genes during different stages of plant development ., However , we identified substantially fewer PcG target genes in the endosperm than previous studies found in seedlings consisting of a mixture of many diverse cell types 11 , 19 as well as in root hair and non-hair specific cell types 31 ., The low number of identified H3K27me3 target genes in endosperm correlates well with reduced expression of the critical PRC2 components MEA and FIS2 in the same tissue 8 , 27 ., A reason for lower expression of PcG proteins and only few PcG protein target genes in endosperm at 1–4 DAP could be that at this time , when mitotic activity is high , the endosperm has not yet acquired its terminal differentiation status 32 ., In contrast , the cells profiled in the other studies 11 , 19 , 31 were mostly fully differentiated ., This is similar to the situation in mammals , where lineage-specific genes often become targeted by PcG proteins only upon cell-fate commitment 33 , leading to cell-type specific PcG target profiles and gene expression patterns 34 , 35 ., Furthermore , it should be noted that the endosperm has fundamentally different developmental origin and fate than vegetative tissues; it is derived after fertilization of the diploid central cell and will not contribute any cells to embryo and the developing new plant ., Therefore , it is also possible that the reduced number of H3K27me3 target genes in the endosperm might reflect a less stringent requirement of PcG-mediated gene regulation in the endosperm than in vegetative tissues ., In the endosperm as well as in vegetative tissues , genes encoding for transcription factors were highly enriched among PcG target genes ( this study and 11 ) , supporting the general idea that PcG proteins regulate cell identity by controlling expression of transcription factors 36 ., Importantly however , H3K27me3 target genes were also prominently enriched for pectinesterases and glycosyl hydrolases - two enzyme classes that degrade major components of plant cell walls 28 , 37 , indicating an important role of the FIS PcG complex in the regulation of endosperm cellularization ., The observed deregulation of both enzyme classes in fis2 mutant seeds might be the underlying cause of endosperm cellularization failure of fis mutants 29 ., Loss of FIS function caused deregulation of only few H3K27me3 genes , similar to observations made in mammalian and Drosophila cells , where only a small subset of PcG target genes were deregulated upon depletion of PcG proteins 33 , 38 , 39 ., Stable repression of FIS target genes could be due to secondary epigenetic modifications that together with FIS-mediated H3K27me3 keep PcG target genes repressed and which are not alleviated in FIS-depleted cells ., Alternatively , it is possible that secondary epigenetic modifications are only recruited to FIS target genes upon loss of FIS function ., As a third and complementary explanation for the lack of expression of a large number of FIS target genes in FIS-depleted endosperm we propose that the promoters of many PcG target genes lack binding sites for endosperm-specific transcriptional activators required for substantially increased expression in this tissue ., This last explanation would imply that those FIS target genes that are deregulated in the fis2 mutant are even in wild type expressed in the endosperm ., Indeed , deregulated FIS target genes were predominantly expressed during wild-type seed development ( Figure 6B ) , supporting the hypothesis that cis-acting tissue-specific enhancers are required for full induction of FIS target genes upon loss of H3K27me3 ., TEs and TEGs were most prominently enriched among endosperm-specific H3K27me3 targets ., This is in contrast to the situation in vegetative tissues , where these elements are largely excluded from PcG target genes 11 ., We propose that reduced levels of DNA methylation in the endosperm allow targeting of the FIS PcG complex to defined sequence elements that are protected by DNA methylation in vegetative tissues ., This conclusion is supported by the following findings made in this study:, ( i ) Shared H3K27me3 targets were completely devoid of DNA methylation , indicating that DNA methylation prevents targeting by PcG proteins ., ( ii ) Endosperm-specific H3K27me3 protein coding genes had much higher CG DNA methylation levels in vegetative tissues compared to genome-wide average DNA methylation levels , supporting the view that DNA methylation prevents these genes being targeted by PcG proteins in vegetative tissues ., ( iii ) In the endosperm , the average DNA methylation level of endosperm-specific H3K27me3 targets was reduced compared to vegetative tissues ., This trend was most pronounced for TEs , where DNA methylation level of endosperm-specific TEs were much lower compared to the genome-wide average DNA methylation of TEs in the endosperm ., However , also TEGs and protein-coding genes had reduced DNA methylation levels in the endosperm compared to vegetative tissues , supporting the notion that reduced DNA methylation levels in the endosperm allow targeting of the FIS PcG complex to defined sequence elements ., However , DNA demethylation is a global phenomenon 13 , 14 , but only selected sequences were targeted by the FIS complex , suggesting that DNA demethylation is necessary , but not sufficient for targeting of the FIS complex ., The conclusion that DNA methylation and H3K27me3 are usually exclusive epigenetic marks is strongly supported by previous studies on seedlings with experimentally altered DNA methylation ., When DNA methylation was reduced , H3K27me3 localized to defined regions within heterochromatin 25 , and when DNA methylation was increased H3K27me3 levels dropped 40 ., Mutual antagonistic placement of DNA methylation and H3K27me3 was also identified at the imprinted Rasgrf1 locus in mouse 41 , suggesting an evolutionary conserved basis of the underlying mechanism ., Together , we conclude that DNA methylation prevents targeting of PcG proteins to sequence elements that have the potential to recruit PcG proteins ., A transgenic Arabidopsis thaliana ( Landsberg erecta ( Ler ) ) line in which endosperm nuclei were specifically marked by EGFP was established by expressing a translational fusion of PHE1 with EGFP under the transcriptional control of the PHE1 promoter ( PHE1::PHE1-EGFP ) and 3 kb regulatory 3′ sequences ., A transgenic Arabidopsis ( Columbia , Col ) line constitutively expressing YFP fused to histone H3 . 2 ( 35S::H3 . 2-YFP ) served as a positive control ., The fis2-1 allele ( Ler accession ) has been described previously 3 ., The met1-3 ( Col accession ) allele was described in 42 ., For met1; fis2 double mutant analysis the newly identified fis2-5 allele ( SALK_009910; Col accession ) was used , containing a T-DNA insertion within the first exon ., The fis2-5 seed abortion ratio and mutant seed phenotypes were analyzed and found to be similar to the fis2-1 allele ( data not shown ) ., Seeds were surface sterilized ( 5% sodium hypochlorite , 0 . 1% Tween-20 ) and plated on MS medium ( MS salts , 1% sucrose , pH 5 . 6 , 0 . 8% bactoagar ) ., Plants were grown in a growth cabinet under long day photoperiods ( 16 h light and 8 h dark ) at 22°C ., After 10 days , seedlings were transferred to soil and plants were grown in a growth chamber at 60% humidity and daily cycles of 16 h light at 22°C and 8 h darkness at 18°C ., Inflorescences were harvested approximately 21 days after transfer to soil , shock-frozen in liquid nitrogen and stored at −80°C ., For analysis of seedlings , seeds were stratified for 2 days at 4°C before incubation in a growth cabinet ., After 10 days , whole seedling tissue was harvested , shock-frozen in liquid nitrogen and stored at −80°C before further usage ., Microscopy imaging was performed using a Leica DM 2500 microscope ( Leica Microsystems , Wetzlar , Germany ) with either bright-field or epifluorescence optics ., Images were captured using a Leica DFC300 FX digital camera , exported using Leica Application Suite Version 2 . 4 . 0 . R1 , and processed using Photoshop 7 . 0 ( Adobe Systems Incorporated , San Jose , USA ) ., Confocal imaging was performed on a Leica SP1-2 ., Nuclei were isolated from 3 . 5 g of inflorescences following the protocol described in 43 ., Isolated nuclei were resuspended in 1× PBS , and proteins were crosslinked to DNA with 1% formaldehyde for 8 min ., After adding glycine to 125 mM final concentration and incubation for 5 min , crosslinked nuclei were washed and resuspended in 1× PBS and stained by addition of Propidium Iodide ( PI ) or DAPI to a final concentration of 1 µg/ml or 0 . 5 µg/ml , respectively ., Biparametric flow analysis of EGFP fluorescence versus nuclear DNA content was performed on a fluorescence activated cell sorter ( FACS Aria II , Becton , Dickinson , Franklin Lakes , USA ) equipped with a 70 µm flow tip and operated at a sheath pressure of 70 psi ., Events were thresholded on forward scatter and samples were sorted at the event rate of 15000/sec ., For EGFP and PI excitation a 488 nm laser and for DAPI excitation a 407 laser were used ., The barrier filters were 610/20 nm for PI , 450/40 for DAPI and 530/30 for EGFP fluorescence ., The position of the nuclei gate was defined using 6 µm beads ( Becton Dickinson ) , forwards ( FSC-A ) and sidewards scatter ( SSC-A ) and was verified by DAPI-staining ( Figure S4A ) ., The position of the sort region was established by first determining the baseline of green fluorescence using inflorescence nuclei from EGFP-negative Ler control plants ( Figure S4B ) ., The upper and left- and right-hand boundaries of the sort window were adjusted to include all nuclei derived from YFP-positive 35S::H3 . 2-YFP control plants ( Figure S4B ) ., Sorted GFP positive nuclei from PHE1::PHE1-EGFP plants were reanalyzed to verify sorting conditions ( Figure S4C ) ., For expression analysis from sorted nuclei , RNA was isolated by flow sorting nuclei directly into 450 µl of RLT lysis buffer ( Qiagen , Hilden , Germany ) and using the RNeasy Plant Mini Kit ( Qiagen ) according to the manufacturers recommendation ., For other expression analyses , siliques were harvested at the indicated time points and RNA extraction and generation of cDNAs were performed using RNeasy Plant Mini Kit ( Qiagen ) according to the suppliers instructions ., For quantitative RT-PCR , RNA was treated with DNaseI and reverse transcribed using the First strand cDNA synthesis kit ( Fermentas , Ontario , Canada ) ., Gene-specific primers and Fast-SYBR-mix ( Applied Biosystems , Carlsbad , USA ) were used on a 7500 Fast Real-Time PCR system ( Applied Biosystems ) ., Analysis was performed using three replicates and results were analyzed as described 44 ., Briefly , mean expression values and standard errors for the reference gene as well as for the target genes were determined , taking into consideration the primer efficiency that was determined for each primer pair used ., Relative expression values we

Introduction, Results, Discussion, Materials and Methods

Polycomb group ( PcG ) proteins act as evolutionary conserved epigenetic mediators of cell identity because they repress transcriptional programs that are not required at particular developmental stages ., Each tissue is likely to have a specific epigenetic profile , which acts as a blueprint for its developmental fate ., A hallmark for Polycomb Repressive Complex 2 ( PRC2 ) activity is trimethylated lysine 27 on histone H3 ( H3K27me3 ) ., In plants , there are distinct PRC2 complexes for vegetative and reproductive development , and it was unknown so far whether these complexes have target gene specificity ., The FERTILIZATION INDEPENDENT SEED ( FIS ) PRC2 complex is specifically expressed in the endosperm and is required for its development; loss of FIS function causes endosperm hyperproliferation and seed abortion ., The endosperm nourishes the embryo , similar to the physiological function of the placenta in mammals ., We established the endosperm H3K27me3 profile and identified specific target genes of the FIS complex with functional roles in endosperm cellularization and chromatin architecture , implicating that distinct PRC2 complexes have a subset of specific target genes ., Importantly , our study revealed that selected transposable elements and protein coding genes are specifically targeted by the FIS PcG complex in the endosperm , whereas these elements and genes are densely marked by DNA methylation in vegetative tissues , suggesting that DNA methylation prevents targeting by PcG proteins in vegetative tissues .

Cell identity is established by the evolutionary conserved Polycomb group ( PcG ) proteins that repress transcriptional programs which are not required at particular developmental stages ., The plant FERTILIZATION INDEPENDENT SEED ( FIS ) PcG complex is specifically expressed in the endosperm where it is essential for normal development ., The endosperm nourishes the embryo , similar to the physiological function of the placenta in mammals ., In this study , we established the cell type–specific epigenome profile of PcG activity in the endosperm ., The endosperm has reduced levels of DNA methylation , and based on our data we propose that PcG proteins are specifically targeted to hypomethylated sequences in the endosperm ., Among these endosperm-specific PcG targets are genes with functional roles in endosperm cellularization and chromatin architecture , implicating a fundamental role of PcG proteins in regulating endosperm development ., Importantly , we identified transposable elements and genes among the specific PcG targets in the endosperm that are densely marked by DNA methylation in vegetative tissues , suggesting an antagonistic placement of DNA methylation and H3K27me3 at defined sequences .

genetics and genomics/genomics, molecular biology/histone modification, genetics and genomics/plant genetics and gene expression, developmental biology/plant growth and development, molecular biology/dna methylation, genetics and genomics/epigenetics, molecular biology/chromatin structure

journal.pgen.1005142

Methylation-Sensitive Expression of a DNA Demethylase Gene Serves As an Epigenetic Rheostat

In plants , animals , and fungi , DNA methylation is used to repress the transcription of potentially harmful DNA sequences 1 ., Targets include long transposable elements ( TEs ) that have intact open reading frames and primarily reside in heterochromatin and shorter TE fragments that are prevalent in euchromatic gene-rich regions ., In plants , DNA methylation is dynamically regulated during development and in response to external perturbations ., Many of these changes occur at TEs or TE-derived sequences ., Examples include modest DNA methylation changes in gene-proximal regions upon exposure to bacteria or bacterial elicitors 2 , 3 and DNA demethylation of TEs in the 5 regions of stress response genes during fungal infection 4 ., Dynamic methylation changes have also been implicated in the regulation of genes in response to abiotic signals 5 , 6 ., Furthermore , DNA methylation is dynamic during reproductive development ., DNA demethylation in the female gametophyte is important for establishing gene imprinting in the endosperm after fertilization 7 ., During male gametogenesis , the sperm become hypomethylated in certain sequence contexts 8 ., Similar to other dynamic changes , the removal of methylation in gametophytes occurs largely at TE fragments in euchromatin ., DNA methylation patterns are a product of methylation and demethylation activities , but how these opposing activities are balanced in the genome is unknown ., In plants , DNA methylation is established and maintained in different cytosine sequence contexts by genetically distinct pathways ., Euchromatic TEs in Arabidopsis and maize are primarily targeted for cytosine methylation through the process of RNA-directed DNA methylation ( RdDM ) , which results in cytosine methylation in all sequence contexts ( CG , CHG , and CHH , where H represents any base other than G ) 1 , 9 ., This process is initiated by transcription of non-coding RNAs by a specialized RNA polymerase unique to plants , RNA Pol IV ., These non-coding RNAs are then converted into dsRNAs by the RNA-dependent RNA polymerase RDR2 ., Small 24 nt RNAs generated from these transcripts are then loaded into AGO4 ., The small RNAs are thought to interact with non-coding transcripts that are generated by a second plant-specific polymerase , RNA Pol V 10 , resulting in the recruitment of the de novo methyltransferase DRM2 and sequence-specific DNA methylation ., 21–22 nt small RNAs generated through an RDR6-dependent pathway can also direct de novo methylation independently of RNA Pol IV 11 , 12 ., Positive feedback between existing and de novo DNA methylation reinforces silencing 13 , 14 ., Maintenance of asymmetric CHH methylation requires continual de novo methylation by RdDM ., Other processes maintain DNA methylation in the CG and CHG sequence context ., CG DNA methylation is maintained by the maintenance methyltransferase MET1 in conjunction with VIM methyl-binding proteins ., CHG methylation is maintained by CMT3 and is positively reinforced by histone H3K9 dimethylation 15 ., By contrast , TEs in heterochromatic sequences are methylated by CMT2 with the assistance of the nucleosome remodeler DDM1 16 , 17 ., Because there is positive feedback between methylation and further RdDM activity and because many RdDM targets are near genes , it is important that mechanisms are in place to protect genes from potentially detrimental hypermethylation ., In the Arabidopsis genome , 44% of genes have a TE within 2 kb of the transcribed region 18 , potentially creating a conflict between TE suppression by RdDM and gene expression ., DNA methylation is opposed by 5-methylcytosine DNA glycosylases that remove methylcytosine from DNA by base excision repair ., Plants with mutations in the three DNA glycosylases expressed in somatic tissues , ROS1 , DML2 , and DML3 , gain methylation in all sequence contexts in gene proximal regions , primarily around TEs and TE-derived sequences 19 , 20 ., DNA demethylation is therefore important to protect genes from RdDM spreading ., This has been demonstrated at several loci ., For example , the EPF2 gene , which is associated with a methylated TE approximately 1 . 5 kb 5’ of its transcriptional start site , gains methylation in the region between the TE and 5’ end of the gene in ros1 dml2 dml3 mutants , resulting in transcriptional silencing 21 ., Although DNA methylation is primarily thought of as repressive to transcription 20 , 22 , expression of the DNA demethylase gene ROS1 is unexpectedly reduced in some DNA methylation mutants 23–26 ., Whether this is a direct or indirect effect has not been demonstrated ., These observations on ROS1 expression form the basis of our study ., Here we describe the existence of a rheostat for genomic methylation activity ., We find that RdDM and DNA demethylation activities converge on TE-derived sequences 5’ of ROS1 ., In contrast to other genomic targets of these pathways , expression of ROS1 is promoted by the RdDM pathway and inhibited by demethylation by ROS1 ., Thus the ROS1 locus functions as a self-regulating epigenetic rheostat , balancing input from both DNA methylation and demethylation to maintain homeostasis between these opposing systems ., Previous studies have shown that ROS1 expression is reduced in mutants in which DNA methylation is disrupted or altered 23–26 ., Additionally , expression of the ROS1 gene is significantly reduced when plants are grown on the methyltransferase inhibitor 5-aza-2-deoxycytidine ( 5-azaC ) 24 ( Fig 1A ) ., Here we systematically evaluated which DNA methylation pathways promote ROS1 expression by performing RT-qPCR on multiple Arabidopsis mutants that directly or indirectly alter DNA methylation ., met1 plants have pleiotropic methylation phenotypes; methylation in CG , CHG and CHH sequence contexts is reduced genome-wide in combination with local regions of non-CG hypermethylation 20 ., We observed extremely low levels of ROS1 transcripts in met1 and vim seedlings ( Fig 1B ) , as has been reported previously 23 , 24 ., An approximately ten-fold decrease in ROS1 transcript levels was observed in eleven different RdDM mutants ( Fig 1C ) , consistent with previous findings that ROS1 expression is reduced in rdr2 , nrpd1a , nrpe1 , drd1 , and drm2 mutants 23 , 25 ., RdDM is predominantly associated with transcriptional repression; therefore transcriptional activation of ROS1 by RdDM potentially represents an under appreciated function for this pathway ., Transcripts from the related 5-methylcytosine DNA glycosylases DML2 and DML3 are present at much lower levels than ROS1 in wild-type tissues ( S1 Fig ) ., Mutations in the RdDM pathway do not alter the transcript abundance of DML2 , and result in small reductions in DML3 ( S1 Fig ) ., No significant changes to ROS1 transcript abundance were observed in CHG methyltransferase mutants ( Fig 1D ) , in mutants of key regulators of histone H3K9 methylation ( Fig 1D ) , which is tightly associated with CHG methylation 27 , in plants with mutations in genes required to establish non-CG methylation in heterochromatin 16 , 17 ( Fig 1E ) , or in plants with a mutation in the RDR6 gene , which can trigger de novo methylation independently of the canonical RdDM pathway 11 , 12 ( Fig 1E ) ., Thus , ROS1 down-regulation in methylation mutants is restricted to mutations in MET1 and its cofactors , and mutations in the RdDM pathway ., To test if ROS1 silencing in met1 or RdDM mutants was heritable , we crossed met1 , rdr2 , and drm1; drm2 plants to wild type and evaluated ROS1 expression in heterozygous F1 progeny ., ROS1 expression remained reduced in MET1/met1 F1 progeny at levels about half that of wild type plants , regardless of whether the met1 plant served as the male or female parent in the cross ( Fig 2A ) ., This suggests that the ROS1 allele inherited from the met1 parent remained silenced through meiosis ., However , ROS1 expression was gradually restored as MET1/met1 progeny developed ( Fig 2B ) ., Thus , erasure of met1-induced epigenetic changes and restoration of normal regulatory mechanisms at ROS1 likely takes place over multiple cell divisions ., By contrast , ROS1 transcripts were restored to wild-type levels in F1 progeny of RdDM mutants crossed to wild type ( Fig 2C ) ., Recently it has been shown that reduced expression of the histone demethylase gene IBM1 contributes to reduced ROS1 expression in met1 mutants 26 ., However , we found that ROS1 expression is not reduced in ibm1 mutants ( Fig 1D ) , nor is IBM1 expression reduced in the RdDM mutants rdr2 and nrpd1a ( RNA Pol IV ) ( S2 Fig ) , suggesting that the decreased expression of ROS1 expression in RdDM mutants is IBM1-independent ., Together these data indicate that the down-regulation of ROS1 observed in met1 and RdDM mutants represents distinct processes , which was further supported by methylation profiling of ROS1 in different mutant backgrounds , described below ., To determine if the ROS1 locus is targeted directly by DNA methylation , we performed bisulfite PCR and sequencing of the entire ROS1 gene and 1 kb of 5’ flanking sequences ( Fig 3 , S3 Fig ) ., In wild-type plants we identified two small regions where cytosines were methylated in CG , CHG , and CHH sequence contexts ( a hallmark of RdDM ) : a 228 bp region partially overlapping an AtREP5 Helitron TE directly upstream of ROS1 , and in sequences encoding exons 15–18 ( Fig 3 ) ., Genome-wide chromatin-IP datasets 10 , 28 showed that peaks of the RdDM proteins NRPE1 ( RNA Pol V ) and AGO4 were present 5 of the ROS1 start codon , overlapping the nearby TE ( Fig 2 ) ., This is the same region where an RNA Pol V transcript has been detected 10 ., There were high levels of CHH methylation in this region , predominantly on the top strand ( Fig 2 , S3 Fig ) , and we identified multiple 24 nucleotide small RNAs directly matching this sequence from published datasets ( Fig 3 ) 29 , 30 ., In addition , we detected multiple small RNAs matching the methylated exons within the ROS1 coding region , but these did not overlap with peaks in the AGO4 or NRPE1 ChIP datasets , consistent with the low levels of CHH methylation in this region ( Fig 3 ) ., To determine the proximity of the 5 methylated region to the ROS1 transcriptional start site ( TSS ) , we performed 5 RACE using RNA from wild-type Col-0 seedlings and identified two transcription start sites , 26 and 442 bp 5’ of the ROS1 start codon ( S3 Fig ) , the latter of which is within 100 bp of the methylated region ., We also profiled ROS1 methylation in met1 and rdr2 mutants ., CG methylation was eliminated in met1 , but the ROS1 coding region was hypermethylated in the CHG context ( Fig 2 ) , as has been previously reported 26 ., We did not observe any evidence for coding region hypermethylation in rdr2 mutants ., Instead , there was a clear reduction in non-CG methylation 5’ of the ROS1 TSS in rdr2 plants , as typically occurs when RdDM activity is lost ( Fig 3 , S3 Fig ) ., Thus the TE at the 5 end of ROS1 is the most likely candidate as the site of RdDM activity that promotes ROS1 expression , despite the fact that methylated TEs 5 of genes are typically associated with transcriptional repression 18 , 20 , 21 ., Combined with the distinct behavior of ROS1 alleles inherited from met1 or rdr2 parents ( Fig 2 ) , we propose that the RdDM pathway acts to promote ROS1 expression via a different mechanism than does the MET1 pathway ., Although our results suggest that ROS1 expression is positively correlated with DNA methylation at the ROS1 locus , reduced expression of ROS1 in RdDM mutants could be direct or indirect , for example due to altered expression of a ROS1 regulator ., To distinguish between these possibilities we sought to restore DNA methylation at the ROS1 5’ region in an rdr2 mutant background and then observe the effect on ROS1 expression ., We attempted to bypass the inability of the rdr2 mutant to make a dsRNA that initiates RdDM by expressing an inverted repeat transgene under the control of the constitutive 35S promoter in rdr2 mutants ( Fig 4A ) ., Transcription of inverted repeat transgenes creates a double-stranded hairpin RNA that can be processed into small RNAs that direct DNA methylation 31 ., We used inverted repeats corresponding precisely to the 228 bp ROS1 5’ region that is methylated in wild type ., We screened rdr2 T1 plants for methylation of the ROS1 5’ region and focused on six lines for in-depth analysis of DNA methylation by bisulfite sequencing ( Fig 4B ) ., Methylation in each of the six lines was restored to varying degrees , constituting an epiallelic series , with some lines exhibiting a methylation profile strikingly similar to wild type ., Methylation occurred in all sequence contexts , indicative of RNA-directed DNA methylation ., ROS1 expression was examined in leaves of each independent line by RT-qPCR ( Fig 4C ) ., Remarkably , ROS1 expression was restored in rdr2 mutants when methylation of the 5’ region was restored ., In a transgenic line with limited restoration of DNA methylation ( line #19 ) , ROS1 expression increased only marginally in rdr2 mutants ( Fig 4C ) ., These data demonstrate that methylation of the 5’ sequence is sufficient to promote ROS1 expression , and eliminate the possibility that decreased expression of ROS1 in rdr2 mutants is caused by an indirect mechanism ., We noticed that cytosines in the CG context 5’ of ROS1 were intermediately methylated at around 50% in wild-type tissues , with independent variance at every CG position in the sequence ( Fig 3 , S3 Fig ) ., CG methylation is faithfully copied by MET1 during DNA replication , and so average methylation at symmetric CG sites is usually close to 0 or 100% 20 ., The observed intermediate level of CG methylation and the low frequency of bisulfite clones fully methylated in the CG context ( S3 Fig ) suggested that active DNA demethylation might also be active at the 5’ sequence ., We hypothesized that ROS1 might oppose RdDM to remove methylation at its own promoter ., We performed bisulfite sequencing of the 5 methylated region in two missense mutants of ROS1 , ros1-2 32 and ros1-7 , an allele encoding a protein with an E956K substitution in the ROS1 DNA glycosylase domain 33 ., Symmetric CG methylation increased to nearly 100% in both mutants compared to their wild-type siblings , along with increases in non-CG methylation ( Fig 5A ) , indicating that ROS1 actively removes methylation from this region in wild-type plants ., At other loci , removal of 5 methylation by ROS1 increases transcription 21 ., We examined the effect of DNA demethylation by ROS1 on ROS1 transcription by performing RT-qPCR on ros1-2 and ros1-7 mutants ., Because these are missense mutations , nonsense-mediated decay should not be a complicating factor in measuring transcript abundance ., ROS1 transcripts were 2 to 4-fold more abundant in ros1 mutants ( Fig 5B ) ., This suggests that active demethylation by ROS1 represses transcription of ROS1 , counteracting the function of the RdDM pathway , which promotes ROS1 expression ., Thus ROS1 regulates the expression of its own gene , forming a negative feedback loop for demethylation activity ., To determine if regulation of ROS1 by methylation might be adaptive , we assessed whether methylation-sensitive expression of ROS1 is conserved in other species ., Arabidopsis lyrata , which diverged from A . thaliana approximately 10 million years ago , has two highly conserved paralogs of ROS1 in tandem in the genome , which we termed AlROS1a and AlROS1b ( Fig 6A ) ., We performed a Bayesian reconstruction of the phylogeny of ROS1 homologs within all sequenced Brassicales ( Fig 6B ) and found that the duplication giving rise to the two ROS1 paralogs in A . lyrata occurred prior to the divergence of A . lyrata from A . thaliana ., AtROS1 belongs to the same clade as AlROS1a , and no true homologs to AlROS1b exist in A . thaliana ( Fig 6B ) ., The homolog to AlROS1b was likely lost in the lineage that gave rise to A . thaliana ., AlROS1a and AlROS1b share a high degree of sequence similarity in their coding region , but no significant similarity in their upstream sequences ., Only AlROS1a has an upstream region conserved with AtROS1 ( Fig 6A ) , including the presence of the same 5’ TE ., The 5’ sequences are 78% identical over the first 1 . 4 kb ., We performed bisulfite sequencing of the AlROS1a 5’ region ( Fig 6C ) and of the exonic sequences in AlROS1a and AlROS1b that match exons 15–18 in ROS1 ( S4 Fig ) ., Although CG methylation was present in the 3’ exonic region of both paralogs , non-CG methylation was absent ( S4 Fig ) ., Additionally , unlike A . thaliana , we did not find any small RNAs matching these exons for either AlROS1a or AlROS1b in a small RNA dataset from A . lyrata flowers 34 ., By contrast , the methylation profile 5’ of AlROS1a was remarkably similar to the methylation profile 5 of AtROS1 ( Fig 6C and 6D and Fig 3 ) ., Like AtROS1 , CG methylation in the 5’ region was at an intermediate level between 40–50% , suggesting that RdDM and active DNA demethylation might also simultaneously target AlROS1a ., Pair-wise alignment of the sequences from each species showed that the methylation was present at conserved cytosines ( Fig 6D ) ., Furthermore , we identified small RNAs from A . lyrata datasets 34 that are almost identical to the small RNAs associated with methylation at the 5 end of AtROS1 ( Fig 6D ) ., The methylated sequence matching these RNAs is directly adjacent to a homolog of the AtREP5 TE upstream of AtROS1 ( Fig 6A ) ., These data indicate that RdDM targeting to a region 5’ of the ROS1 TSS is evolutionarily conserved ., To test whether the expression of either AlROS1 paralog was responsive to methylation alterations , A . lyrata seedlings were grown on varying concentrations of 5-azaC ., AlROS1a but not AlROS1b transcripts were significantly decreased in seedlings grown on 5-azaC ( Fig 6E and 6F ) ., Thus expression of the true A . lyrata homolog of ROS1 , AlROS1a , which has similar 5’ methylation and conserved small RNAs , is methylation-responsive ., Together , these data suggest that the regulation of ROS1 by RdDM and DNA demethylation at 5’ sequences is conserved between A . thaliana and A . lyrata ., Interestingly , in transcriptome datasets from shoot apical meristems or immature ears of Z . mays mop1 mutants ( Mop1 is an RDR2 homologue ) , expression of two DNA glycosylase genes with high homology to AtROS1 ( DNG101 and DNG103 ) was reduced 2 to 3 . 3-fold in comparison to wild type 35 , 36 ., Reduced expression of ROS1 homologs has also been observed in the transcriptome of Z . mays RNA polymerase IV mutants 37 ., We confirmed that DNA glycosylase expression is reduced in mop1 leaves by RT-qPCR ( S5 Fig ) ., Both DNG101 and DNG103 have methylated TEs in their 5 region in all sequence contexts , suggesting that both loci could be direct targets of RdDM 38 ., These data further suggest that regulation of DNA glycosylases by RdDM might be a general feature of angiosperms , and thus likely adaptive ., Our data demonstrate that ROS1 functions as a self-regulating epigenetic rheostat ., RdDM and DNA demethylation activities converge at the ROS1 locus , but each activity has the opposite outcome on ROS1 transcript abundance as compared to typical targets of these processes ( Fig 7 ) ., By establishing DNA methylation at the ROS1 locus in rdr2 mutants ( Fig 4 ) , we have conclusively shown that methylation of the 5’ sequence is sufficient to restore ROS1 expression ., Thus reduced expression of ROS1 in rdr2 mutants is not caused by an indirect mechanism , such as decreased expression of another gene required for ROS1 expression or increased expression of a negative regulator of ROS1 ., While the precise mechanism by which the activity of ROS1 is repressive and RdDM is activating remains unknown , we speculate that a protein that either negatively or positively regulates ROS1 may exhibit differential DNA binding affinity based on methylation of the underlying ROS1 5’ DNA sequence ., Alternatively , it is possible that rather than DNA methylation itself , the act of RdDM could play a regulatory role ., For example , occupancy , DNA melting or elongation by RNA polymerases IV or V could be required for positive regulatory factors to access ROS1 ., Interestingly , RdDM in an intron of the MADS3 gene in Petunia has also been shown to be associated with transcriptional upregulation 39 ., In this instance , it is likely that methylation of a short cis-element is necessary to confer increased expression ., The reversal of methylation outcomes at ROS1 permits the genome to maintain gene expression ( promoted by DNA demethylation ) and genome defense ( TE silencing by RdDM ) in homeostatic balance ., For example , if the genome were under stress from high TE transcriptional activity or invasion , RdDM activity might increase ., Under these conditions , RdDM activity would promote the expression of ROS1 , so that the activity of demethylation at target genes would be maintained in equilibrium with the activity of RdDM in the genome ( Fig 7 ) ., Conversely , if RdDM were less active , ROS1 expression would be reduced to prevent hypomethylation as a consequence of demethylation activity ., The result is that the expression of ROS1 is always maintained in balance by its autoregulation ( Fig 7 ) , which may help underpin the regulation of epigenetic homeostasis within plants and explain why spontaneous changes to methylation are generally very rare 40 , 41 ., This homeostatic balance may be dynamically modified in certain conditions , such as fungal or bacterial infections , where active demethylation of defense genes is important for resistance 2–4 ., The rheostat may also be important during normal development ., In pollen , DRM2 is expressed at a low level in microspores and sperm 8 ., In agreement with our findings from drm2 and other RdDM mutants , ROS1 transcripts have not been detected in wild-type sperm 42 ., Published methylation profiling data from pollen show that the ROS1 5’ region is hypomethylated in sperm cells , but not in the vegetative nucleus ( S6 Fig ) , where both DRM2 and ROS1 are expressed 42 , 43 ., In the future , it will be important to determine whether disrupting the rheostat has consequences for the proper establishment of methylation patterns in sperm or in developing progeny after fertilization ., In addition to the methylation-responsive regulation of ROS1 that we have described here , it is possible that additional mechanisms underlie epigenetic homeostasis in Arabidopsis ., Targeting of RdDM , H3K9 methylation , and H3K27 methylation are redirected throughout the genome in met1 mutants 20 , 24 , 44 , and it has been hypothesized that this redirection may be the result of compensatory mechanisms necessary to maintain some level of integrity in gene expression 24 ., Therefore , other mechanisms may also regulate repressive histone modifications in balance with gene expression ., The role of methylation in promoting IBM1 expression 26 may be one such example ., Although our experiments have focused on Arabidopsis species , the concept of epigenetic homeostasis might also apply more broadly to other DNA methylation systems , including those in mammals ., It is known that the cancer epigenome exhibits global DNA hypomethylation and local hypermethylation 45 , which is broadly similar to the methylation phenotype of a met1 mutant ., Interestingly , expression of the three TET enzymes , which are responsible for initiating DNA demethylation in mammals by oxidizing 5-methylcytosine , is reduced in multiple cancers 46 ., We conclude that the ROS1 locus serves as a rheostat for methylation levels ., We propose that the ROS1 epigenetic rheostat evolved to counter-balance positive feedback between DNA methylation and RdDM activity 13 , 14 to prevent ectopic gain of DNA methylation ., The conservation of methylation-sensitive ROS1 expression among divergent angiosperms suggests that this regulation is adaptive and could underpin how plants balance a number extremely effective , potent , and self-reinforcing silencing mechanisms while maintaining gene transcription ., Arabidopsis thaliana , Arabidopsis lyrata , and Zea mays plants were grown in a greenhouse with 16-hour days at 21°C ., For experiments performed on whole seedlings , plants were grown on 0 . 5 x MS medium with 5% agar ., For treatment with 5-azaC , A . thaliana or A . lyrata seedlings were grown on filter paper moistened with water or 5 , 10 , 15 or 20 μM 5-aza-2-deoxycytidine ., Fresh water or 5-azaC was added daily ., The accession number for all mutant plants used in this study are in S1 Text ., Total RNA was extracted using an RNeasy Plant Mini Kit ( Qiagen ) ., RNA was extracted from whole 7-day old Arabidopsis thaliana seedlings for all experiments , except for experiments using 5-azaC , in which case 5-day old A . thaliana or A . lyrata seedlings were used , or experiments with transgenic lines expressing a ROS1 inverted repeat construct , in which case rosette leaves from 21-day old plants were used ., RNA was extracted from the tip of the third true leaf of maize plants ., For each genotype , 3 biological replicates of at least 5 pooled individual seedlings ( Arabidopsis ) or individual plants ( maize ) were collected ., Genomic DNA was removed using amplification-grade DNAseI ( Invitrogen ) ., cDNA was synthesized from 500 ng RNA using Superscript II reverse transcriptase ( Invitrogen ) according to manufacturers’ instructions , selecting for polyadenylated transcripts using an oligo-dT primer ., For each cDNA synthesis reaction , a control was performed without addition of reverse transcriptase to test the efficacy of the DNAse treatment ., Quantitative RT-PCR ( RT-qPCR ) was performed using Fast Sybr-Green mix ( Applied Biosystems ) according to manufacturers’ instructions ., All reactions were performed using a StepOne Plus Real-Time PCR system ( Applied Biosystems ) ., Primers were designed to have matching melting temperatures between 60–65 °C and to produce amplicons between 80–160 bp in length ., All primers were used in a final concentration of 400 nM ., The efficiency of all primer pairs was verified using a standard curve dilution of template cDNA prior to their use in quantification of transcripts ., Melt curves were analyzed to verify the presence of one amplicon in each reaction , and representative products were also verified by agarose gel electrophoresis ., Relative expression was calculated using the ddCt method as described 47 ., For Arabidopsis , the reference transcript used for all reactions was AT1G58050 , experimentally verified to be one of the most consistently abundant transcripts in A . thaliana 48 ., For maize , the reference transcript was ZmEF1α , defined to be the most consistent reference transcript over the majority of experimental conditions 49 ., Primer sequences are available in the S1 Text ., 5 RACE of ROS1 was performed using 10 μg Col-0 RNA extracted from 10-day old seedlings ., The 5 RACE cDNA library was synthesized using a FIRST-CHOICE RLM-RACE Kit ( Ambion ) according to manufacturers’ instructions , with the exception that a ROS1-specific oligonucleotide was used to prime cDNA synthesis ., RACE products were amplified using a nested PCR strategy , purified using a QIAquick gel extraction kit ( Qiagen ) and cloned for sequencing using a TOPO-Blunt PCR cloning kit ( Invitrogen ) ., Genomic DNA was isolated from 7-day old seedlings or 21-day old rosette leaves using a CTAB procedure ., 2 μg DNA were sheared by sonication and used for bisulfite treatment , which was performed as described 50 ., 2 μl bisulfite treated DNA was used in PCR reactions with 2 . 5 U ExTaq DNA polymerase ( Takara ) and 0 . 4 μM primers using the following cycling conditions ( 95 °C 3 minutes , 40 cycles of 95 °C for 15 seconds , 52 °C for 60 seconds , 72 °C for 60 seconds , 72 °C for 10 minutes ) ., PCR products were cloned using TOPO-TA ( Invitrogen ) or CloneJet ( Life Technologies ) PCR cloning kit and individual colonies were sequenced ., Sequenced products were aligned using MUSCLE 51 , and methylation of each cytosine residue was calculated using CyMate 52 ., The 228 bp sequence 5 of ROS1 that is targeted by RdDM in wild-type plants was amplified and cloned into the directional entry vector pENTR-TOPO-D ( Invitrogen ) ., The sequence was then inserted twice in an inverted repeat conformation into the vector pANDA-35HK using a single LR clonase reaction as described by 53 ., rdr2 mutant plants were transformed with the inverted repeat transgene by floral dipping 54 , and T1 lines were screened for DNA methylation 5 of ROS1 using a restriction enzyme assay on bisulfite treated DNA ., 90% of lines screened exhibited higher DNA methylation than rdr2 and nrpe1 mutants ., Six lines covering a range of methylated epigenotypes were selected for bisulfite sequencing and expression analysis ., Coding sequences for ROS1 homologs from all fully sequenced genomes within Brassicales were downloaded from Phytozome 9 . 1 ., In addition , the coding sequence for Arabidopsis thaliana DME , which belongs to a distinct clade of DNA glycosylases 55 , was included as an out-group ., Sequences were aligned using Muscle 3 . 8 51 , and manually verified using Aliview 56 ., The phylogeny was reconstructed from the aligned matrix using MrBayes 3 . 1 . 2 57 ., jModelTest 58 was used to determine the ideal model for analysis , and the general time reversible model with gamma distributed rate variation was chosen ., Gaps were treated as missing data ., The analysis was run for 2 , 500 , 000 generations sampling every 100 trees ., By this time the average standard deviation of split frequencies had reached <0 . 01 and the potential scale reduction factor ( PRSF ) was <1 . 005 for all parameters ., The first 25% of trees were discarded as burn-in and the output phylogeny was further analyzed and annotated using FigTree 1 . 4 .

Introduction, Results, Discussion, Materials and Methods

Genomes must balance active suppression of transposable elements ( TEs ) with the need to maintain gene expression ., In Arabidopsis , euchromatic TEs are targeted by RNA-directed DNA methylation ( RdDM ) ., Conversely , active DNA demethylation prevents accumulation of methylation at genes proximal to these TEs ., It is unknown how a cellular balance between methylation and demethylation activities is achieved ., Here we show that both RdDM and DNA demethylation are highly active at a TE proximal to the major DNA demethylase gene ROS1 ., Unexpectedly , and in contrast to most other genomic targets , expression of ROS1 is promoted by DNA methylation and antagonized by DNA demethylation ., We demonstrate that inducing methylation in the ROS1 proximal region is sufficient to restore ROS1 expression in an RdDM mutant ., Additionally , methylation-sensitive expression of ROS1 is conserved in other species , suggesting it is adaptive ., We propose that the ROS1 locus functions as an epigenetic rheostat , tuning the level of demethylase activity in response to methylation alterations , thus ensuring epigenomic stability .

Organisms must adapt to dynamic and variable internal and external environments ., Maintaining homeostasis in core biological processes is crucial to minimizing the deleterious consequences of environmental fluctuations ., Genomes are also dynamic and variable , and must be robust against stresses , including the invasion of genomic parasites , such as transposable elements ( TEs ) ., In this work we present the discovery of an epigenetic rheostat in plants that maintains homeostasis in levels of DNA methylation ., DNA methylation typically silences transcription of TEs ., Because there is positive feedback between existing and de novo DNA methylation , it is critical that methylation is not allowed to spread and potentially silence transcription of genes ., To maintain homeostasis , methylation promotes the production of a demethylase enzyme that removes methylation from gene-proximal regions ., The demethylation of genes is therefore always maintained in concert with the levels of methylation suppressing TEs ., In addition , this DNA demethylating enzyme also represses its own production in a negative feedback loop ., Together , these feedback mechanisms shed new light on how the conflict between gene expression and genome defense is maintained in homeostasis ., The presence of this rheostat in multiple species suggests it is an evolutionary conserved adaptation .

journal.pcbi.1005435

Variable habitat conditions drive species covariation in the human microbiota

Species in an ecosystem interact with each other and with their environment ., Both types of interactions leave an imprint on the composition and diversity of a community ., Two species competing for exactly the same resources engage in a struggle for existence 1 ., In the end , one species will win the competition by driving the other to extinction ., As a result , one might expect that closely related species rarely occupy the same habitat where they would risk being drawn into competition ., On the other hand , species that survive in the same habitat must share many common features 2 ., Thus , the rise and fall of a common resource may cause the abundances of similar species to rise and fall in concert ., These opposing ecological forces simultaneously push and pull on species abundances to shape the composition of a community ., Ecological processes operate on the thousands of microbial species that inhabit the human body 3–6 just as they operate on the Amazon rainforest ., Technological advances have recently made it possible to study the human microbiota using 16S ribosomal RNA tag-sequencing and whole genome ‘shotgun’ metagenomics 7 ., Variability in the composition of the microbiota can be studied in two ways ., Longitudinal studies follow the relative abundances of the species in a single bodysite of a particular person over time 8 ., Cross-sectional studies examine the relative abundances of the species in a single bodysite across a sample of many different people 9 ., These studies have demonstrated that the composition of the human microbiota exhibits three qualitative scales of variation 10 , 11: there are small-scale fluctuations in relative species abundances through time , there are medium-scale variations in species composition from person-to-person , and there are large-scale differences in species composition between different bodysites ., In this work we quantitatively explore the idea that variations in species abundances between different sites can be explained mainly through the local variations in resource availability ., Concretely , we reanalyze data from the Human Microbiome Project ( HMP ) 3–5 , 12 on the species composition of different bodysites ( i . e . , gut , skin , vagina , and oral cavity; Fig 1 ) ., To develop the analysis method , we start from a theoretical model that assumes maximal diversity of species and derives a relationship between species abundance and resource availability ., We then use the results of this model to guide the joint analysis of the datasets from different body sites in the HMP project , introducing a new Common Component Analysis ( CoCA ) method ., The general idea is that if the same species exist in different body sites , the same resources must also be present at these body sites ., We find that this intuition is correct by showing that the covariance of species abundances at specific body sites can be simultaneously projected into the same basis that describes the availability of effective resources ., This means that the abundances of species at different body sites are driven by the same set of resources , just different resources are of varying importance at different body sites ., These results cannot be reproduced from randomized data and reflect an underlying global set of ecological resources shared between body sites ., Our method identifies species that share common resources ., To understand the source of this sharing , we look for similarities between the species ., We find that species that share common resources are also closely related taxonomically , suggesting that evolution is constrained by ecology ., We further back this observation by identifying specific metabolic pathways that are conserved between species identified as close using our resource sharing analysis ., The goal of this work is three-fold ., First , we introduce a new analysis technique that identifies covariations among components in different subsets that are driven by the same process ( CoCA ) ., Second , by successfully applying the CoCA analysis technique to HMP data , we show that the diversity of the microbiome at different body sites is shaped by common biological processes ., Third , in the case of the specific problem of species at different ecological sites , we make the biological point that microbiome species that share the same resources are also closely related taxonomically and we back this fact by identifying shared pathways ., The methodological developments are presented in the “Theory” subsection of the Results and the biological results in the“Analysis” subsection ., For readers predominantly interested in the biological results the “Analysis” subsection can be read independently of the “Theory” subsection ., In addition to presenting statistical evidence for the observations described above , we formulate a hypothesis ( in the form of a mathematical model ) that explains their origin ., Our model is inspired by MacArthur’s famous model of competition 1 , 13–15 , but is adapted to account for the compositional nature of metagenomic survey data ., We demonstrate that habitat variability is sufficient to explain the medium-scale variations in species composition observed in a cross-sectional study of the human microbiota ., As a result , the relative abundances of closely related species are positively correlated—they rise and fall in concert as habitat conditions vary from person-to-person ., Therefore , cross-sectional studies allow us to extract a wealth of information about the influence of species traits and habitat properties on community composition using advanced statistical techniques ., Although we have stated the maximum diversity hypothesis as a generative model of species composition , we rarely know , and generally cannot measure , all of the effective resources in a community ., Therefore , we treat the mathematical model as an inverse problem with the goal of inferring the effective resources from observations of species composition across many individuals and bodysites ., The inverse problem can be solved because , by construction , the model imposes that the inferred effective resources correspond to directions with high intra-bodysite variability ( Fig 2A–2C ) ., We exploit this feature to developed a technique we called CoCA that infers the characteristics of the species and habitats from observed correlations ( S1 Text ) ., Like other techniques for simultaneous matrix diagonalization 19–21 , 28 , CoCA aims to find a single set of directions that simultaneously explain variation within each of the bodysites ., Moreover , CoCA has a theoretical interpretation derived from the maximum diversity hypothesis and properly accounts for the compositional nature of genomic survey data ., Our statistical analysis of the data from the Human Microbiome Project ( HMP ) centers on three observations: The first point is a validation of our mathematical model , whereas points 2 and 3 demonstrate that the results obtained by CoCA have biologically reasonable interpretations ., Point 3 is a corollary of point 2; if taxonomy explains species relationships in the common basis then it follows that there will also be a relationship with characteristics that vary by taxonomy ( e . g . , metabolic pathways ) ., Nevertheless , it is important to check that the pathways that are selected make biological sense ., We compare the CoCA results to PCA , as well as to the results of the CoCA algorithm applied to randomized data ( S1 Text ) ., Previous studies have revealed that bacteria exhibit tremendous genomic and functional diversity due , in part , to high rates of horizontal gene transfer ( HGT ) 36 ., As a result , the ability of sequence-based or taxonomic classification of bacteria to capture ecological relationships has been called into question 37–40 ., Nevertheless , we found that genetically related species respond to fluctuating habitat conditions in the same way , implying that they occupy similar ecological niches ., Thus , current taxonomic groupings of bacteria are largely sufficient for explaining cross-sectional correlations in relative species abundances over the healthy human population ., This result is not at odds with high rates of HGT; it simply implies ecologically derived constraints on evolution ., We introduced CoCA , a theory-driven data analysis technique that can be applied to any cross-sectional study with labeled metadata , including studies with populations corresponding to healthy and unhealthy individuals ., Although the effective resources identified by CoCA are derived entirely from data on relative species abundances across a population , they reflect indirect ecological relationships between species that are mediated through resources and form the basis of a common resource sharing niche space of the human microbiota ., Future analyses of larger , and more diverse , datasets will further elucidate the relationship between this underlying niche space and the functional properties of the organisms in the microbiota ., Given that CoCA identifies features that separate the bodysites with high fidelity , we believe that it is a useful technique for identifying microbiota based biomarkers that discriminate between host phenotypes ., Extending our results to include data from unhealthy subjects will be an important avenue for future work ., A recent paper by Bashan et al 41 developed an approach to analyzing microbial dynamics based on a Dissimilarity-Overlap Curve ., They found that communities with a high overlap in the species that were present also have a low dissimilarity in their relative abundance profiles ., They argue that this relationship is evidence of “universality” where interspecies interactions are essentially the same across a population of human subjects ., Our model is also based on the assumption that the underlying drivers of variation in the microbiota are the same across subjects and across bodysites , and it is only the relative importance of these factors that leads to differences between groups ., However , we focused only on variation in the relative abundances of highly abundant species that are present across all four major bodysites in the Human Microbiome Project rather than variation in species assemblages ., The successful application of CoCA to HMP data from four different body sites implies that the processes that shape the variation in species abundances are shared between bodysites , and only changes in the specific contribution of various effective resources differentiate bodysites ., CoCA uses simultaneous diagonalization to identify processes that are shared between communities ., Covariance matrices from communities without shared drivers of variation cannot be simultaneously diagonalized , as we showed with randomized data ., Consequently , we would expect that CoCA would fail on datasets from clearly different ecological environments ( e . g . , hot springs compared to body sites ) ., In this case , failure is not a bad thing: it can be easily diagnosed from the poor agreement between the predicted and observed covariances and it provides an ecologically meaningful result be ruling out the hypothesis that the environments have shared drivers of variation ., CoCA does not explain 100% of the variation in the HMP data , nor do taxonomic relationships explain 100% of the variation in the inferred resource utilizations ., The additional variation is likely do to other types of interactions between species in the human microbiota that cannot be captured using effective resources that are shared across bodysites ., Moreover , effective resources are only defined by a statistical model and , therefore , do not have obvious relationships to measurable environmental variables ., Here , we attempted to explain the inferred effective resources in terms of metabolic processes inferred through KEGG pathways but it is likely that other factors , such as resilience to temperature or pH ranges , contribute to the effective resources in ways that our analysis with KEGG pathways could not uncover ., Our study also has other limitations that should be addressed in future work ., We have based our analyses on relative species abundances derived from OTUs constructed using data from the HMP ., These data are likely to be noisy , but the degree of uncertainty is difficult to quantify ., Moreover , the use of OTUs defined by 97% sequence identity , and subsequent reduction of the communities to 100 highly abundant species , leads to a coarse grained representation that may smooth out relevant features ., It will be important to revisit our analyses on additional datasets , and with additional tools for generating highly accurate pictures of community composition ., On the theoretical side , it will be important to examine the validity of the maximum diversity hypothesis across communities with different properties ., We analyzed data from the Human Microbiome Project ( HMP ) on person-to-person variability in relative species abundances in four bodysites ( gut , oral cavity , vagina , and skin; Fig 1A ) 3–5 , 12 ., The species-level relative abundances derived from the HMP whole genome sequencing data were obtained from MG-RAST ( Project 385 ) through the MR-RAST API 42 ., Only the processed data as provided on the MG-RAST server were extracted ., Thus , these species counts were constructed using the default MG-RAST pipeline 43 ., Briefly , this pipeline identifies putative rRNA fragments and clusters them at 97% identity to define operational taxonomic units , which are assigned species labels using a search against the M5rna database 44 ., We eliminated any lowly abundant species and selected for further study 100 species ( Fig 3A ) that were highly abundant across all bodysites , as described in S1 Text ., The final dataset ( consisting of the counts of the 100 selected species in each of the samples ) is available in the Supporting Information ( S1 Code ) and at https://sites . google . com/site/charleskennethfisher/home/programs-and-data along with the source code ., Log-ratio transformations are obtained using y = G log x , where x is an N dimensional vector of relative abundances , G is an N − 1 × N matrix with G1 = 0 , and y is an N − 1 dimensional vector of transformed relative abundances ., We use a G that implements an additive log-ratio ( or ALR ) transform , but the choice of G is not critical for our analyses and some other possible choices are discussed further in the S1 Text ., Applying a log-ratio transformation to Eq 2 gives y = G V λ = V ˜ λ ., Here , V is an N × N − 1 dimensional matrix whereas V ˜ = G V is an N − 1 × N − 1 dimensional matrix ., Once again , the use of relative abundances shows up as a loss of information ., The matrix V that contains the information about all N species that we would like to obtain can only be recovered from V ˜ using some assumptions ., The use of compositional transformations with CoCA requires an extra step to recover the N × N − 1 dimensional matrix V from the N − 1 × N − 1 dimensional matrix V ˜ = G V . Unfortunately , the matrix G is not invertible ., But , if we assume that V is sparse then it is possible to determine V from V ˜ ., In the context of the model , this assumption means that any individual consumer species is unlikely to be able to utilize every effective resource ., To recover V from V ˜ , we solve the problem:, min | | V | | 1 subject to G V = V ˜ ( 5 ), where ||V||1 = ∑iμ |Viμ| ., Using the ALR transform , all of the solutions to this problem are all of the form V i μ = z μ + V ˜ ( i - 1 ) , μ ( 1 - δ i 1 ) for i = 1 , … , N , where zμ can , in principle , take on any real value ., Because we want the solution with a minimum L1 norm , it is sufficient to test zμ = 0 and z μ ∈ { - V ˜ i , μ } i = 1 N - 1 ( the only sparse solutions ) and to choose the one with minimum norm ., This is a tractable search over N ( N − 1 ) possibilities in the worst case and can be done easily for reasonable system sizes ., Each row of the matrix V = G V ˜ ( here , G is a matrix that arises from the log-ratio transform—see details ) describes how one of the species responds to changes in the latent variables ., Thus , the ith row of V is a mathematical representation of species i ., The distance between species i and j in the inferred basis can be calculated by computing the distance between the ith and jth rows of V with each column ( i . e . , latent variable ) weighted by its variance ( S1 Text ) ., Distances between species computed from the common components were regressed against the distances computed from KEGG pathways ( S1 Text ) ., To select relevant pathways , we compute posterior probabilities for each regression coefficient to be non-zero using the Bayesian Ising Approximation ( BIA ) 34 , 35 ., The BIA approximates the posterior distribution of a vector indicator variables with si = +1 if pathway i relevant and si = − 1 if pathway i is not relevant ., The posterior distribution is approximately an Ising model described by:, log P λ ( s | y ) ≃ n 2 4 λ ∑ i h i ( λ ) s i + 1 2 ∑ i , j ; i ≠ j J i j ( λ ) s i s j ( 6 ), where the external fields ( hi ) and couplings ( Jij ) are defined as:, h i ( λ ) = r 2 ( y , x i ) - 1 n + ∑ j J i j ( λ ) ( 7 ) J i j ( λ ) = λ - 1 r 2 ( x i , x j ) - n λ r ( x i , x j ) r ( y , x i ) r ( y , x j ) - 1 2 r 2 ( y , x i ) r 2 ( y , x j ) ( 8 ), and λ is the inverse variance of the prior distribution ., Here , r ( z1 , z2 ) is the Pearson correlation coefficient between variables z1 and z2 ., The BIA approximation is based on a series expansion that is valid as long as:, λ ≥ λ * = n ( 1 + p r ) ., ( 9 ), where r = p - 1 ( p - 1 ) - 1 ∑ i ≠ j r 2 ( X i , X j ) is the root mean square correlation between features ., To perform feature selection , we are interested in computing marginal probabilities Pλ ( sj = 1|y ) ≃ ( 1 + mj ( λ ) ) /2 , where we have defined the magnetizations mj ( λ ) = 〈sj〉 ., While there are many techniques for calculating the magnetizations of an Ising model , we focus on the mean field approximation which leads to a self-consistent equation: This mean field approximation provides a computationally efficient tool that approximates Bayesian feature selection for linear regression .

Introduction, Results and discussion, Methods

Two species with similar resource requirements respond in a characteristic way to variations in their habitat—their abundances rise and fall in concert ., We use this idea to learn how bacterial populations in the microbiota respond to habitat conditions that vary from person-to-person across the human population ., Our mathematical framework shows that habitat fluctuations are sufficient for explaining intra-bodysite correlations in relative species abundances from the Human Microbiome Project ., We explicitly show that the relative abundances of closely related species are positively correlated and can be predicted from taxonomic relationships ., We identify a small set of functional pathways related to metabolism and maintenance of the cell wall that form the basis of a common resource sharing niche space of the human microbiota .

The human body is inhabited by a vast number of microorganisms comprising the human microbiota ., The species composition of the microbiota varies considerably from person-to-person and the relative abundances of some species rise and fall in concert ., We introduce a mathematical model where differences in habitat conditions cause most of the variability of the microbiota ., A statistical analysis shows that variable habitat conditions are sufficient for explaining the patterns of variation observed across a healthy human population and , as a result , the correlation between the relative abundances of two species reflects how closely related they are rather than how they directly interact with each other .

taxonomy, ecology and environmental sciences, ecological niches, microbiome, microbiology, random variables, covariance, multivariate analysis, data management, mathematics, statistics (mathematics), algebra, microbial genomics, research and analysis methods, computer and information sciences, medical microbiology, mathematical and statistical techniques, principal component analysis, vector spaces, probability theory, microbial ecology, community ecology, linear algebra, ecology, genetics, biology and life sciences, physical sciences, genomics, statistical methods

journal.ppat.1003926

Shigella Type III Secretion Protein MxiI Is Recognized by Naip2 to Induce Nlrc4 Inflammasome Activation Independently of Pkcδ

Recognition of intracellular pathogenic bacteria by members of the nucleotide-binding domain and leucine-rich repeat containing ( NLR ) family triggers immune responses against bacterial infection 1 , 2 ., A major response against several pathogenic Gram-negative bacteria , including Salmonella , Legionella , and Shigella is the activation of caspase-1 via Nlrc4 in macrophages 1 , 3 ., Upon bacterial stimulation , Nlrc4 mediates the formation of a multi-protein complex termed the inflammasome that induces the activation of caspase-1 leading to the proteolytic maturation of pro-IL-1β and pro-IL-18 as well as the induction of pyroptotic cell death in macrophages 4–6 ., Many Gram-negative bacteria encode a type III secretion system ( T3SS ) with conserved structural features that promote virulence by injecting bacterial effector proteins directly into the cytosol of host cells 7 , 8 ., In macrophages infected with Salmonella , the cytosolic delivery of flagellin or the bacterial rod protein PrgJ through the T3SS is recognized by Nlrc4 leading to inflammasome activation 9 ., Recently , Naips ( NLR family , apoptosis inhibitory proteins ) have been shown to act as adaptor molecules that connect flagellin or the bacterial rod protein PrgJ to Nlrc4 10 , 11 ., Specifically , Naip5 and Naip6 associate with flagellin to promote Nlrc4 oligomerization and inflammasome activation , whereas Naip2 links PrgJ to Nlrc4 10–12 ., These findings suggest a model in which certain Naips specifically recognize flagellin or PrgJ to mediate Nlrc4 inflammasome activation ., Recent studies , however , have revealed that the activation of Nlrc4 is more complex in that phosphorylation of Nlrc4 at Ser533 was found to be critical for the activation of the inflammasome 13 ., Furthermore , it was suggested that Pkcδ is the major Nlrc4 kinase responsible for Nlrc4 phosphorylation and inflammasome activation 13 ., Shigella are non-flagellated bacterial pathogens that contain highly evolved invasion systems that enable them to invade host cells and colonize the epithelium of the large intestine , which ultimately leads to a severe form of colitis called bacillary dysentery 14 ., After uptake of Shigella by intestinal macrophages , the bacterium delivers a subset of effector proteins via the T3SS apparatus into the host cytosol 7 , 8 , 15 ., The inner rod of the T3SS needle complex forms a conduit for protein transport through the periplasm which is assembled by the polymerization of PrgJ in Salmonella and its homologue MxiI in Shigella 16 , 17 ., Because of the homology of Salmonella PrgJ with Shigella MxiI , it can be predicted that Shigella induces activation of Nlrc4 via the sensing of MxiI by host macrophages ., Consistent with this notion , the T3SS of Shigella is required to induce IL-1β secretion and pyroptosis via the Nlrc4 inflammasome 18 ., Furthermore , ectopic expression of MxiI reduced the viability of macrophages and this was inhibited in the absence of Nlrc4 9 ., However , the mechanism by which Shigella MxiI induces activation of the Nlrc4 inflammasome remains unknown ., In this study , we provide evidence that MxiI mediates the activation of the Nlrc4 inflammasome through interactions with Naip2 ., Furthermore , we demonstrate that Naip2 , but not Naip5 , is critical for the interaction of MxiI with Nlrc4 and the activation of the inflammasome in macrophages infected with Shigella ., Finally , we show that Pkcδ is dispensable for Nlrc4 activation ., In the case of flagellated pathogenic bacteria , flagellin is a major and potent stimulator of the Nlrc4 inflammasome ., In addition , Salmonella T3SS rod protein PrgJ is sensed by Nlrc4 to activate caspase-1 ., Because Shigella are unflagellated bacteria , we hypothesized that the Shigella T3SS rod protein MxiI , a homologue of Salmonella PrgJ , induces the activation of the Nlrc4 inflammasome ., To test this hypothesis , we expressed MxiI in wild-type ( WT ) and Nlrc4-deficient bone marrow-derived macrophages ( BMDM ) using a MSCV-IRES-GFP retroviral vector and assessed cell viability by the numbers of viable green fluorescence protein ( GFP ) -positive cells ., After overnight culture , the viability of WT macrophages was dramatically decreased by MxiI-GFP expression when compared to expression of GFP ( Figure 1A ) ., Importantly , the decrease in cell viability was inhibited in Nlrc4−/− macrophages ( Figure 1A ) ., Consistently , expression of MxiI-GFP , but not GFP , induced the release of IL-1β in WT macrophages , which was abolished in macrophages lacking Nlrc4 ( Figure 1B ) ., These results indicate that expression of MxiI induces the activation of the Nlrc4 inflammasome ., We next tested whether the rod protein MxiI interacts with Naip2 or Naip5 in macrophages ., Because expression of MxiI in macrophages causes cell death ( Figure 1A ) , we used macrophages from caspase-1-deficient mice to assess the interaction of MxiI with Naip proteins by immunoprecipitation ., In these experiments , we expressed T7-tagged MxiI in the presence of HA-tagged Naip2 , HA-tagged Naip5 or control plasmid ., Immunoprecipitation analysis showed that MxiI associated with Naip2 , but much less with Naip5 as revealed by immunoblotting with anti-HA antibody ( Figure 2A ) ., Next , we investigated the interaction between Nlrc4 and Naip2 in Shigella-infected macrophages ., To assess this , we expressed T7-tagged Nlrc4 and HA-tagged Naip2 or Naip5 , or control empty vector in uninfected or caspase-1-deficient macrophages infected with WT or an isogenic Shigella strain deficient in the T3SS ( S325 ) ., Immunoprecipitation analysis revealed that Naip2 interacts with Nlrc4 in macrophages infected with WT Shigella ( Figure 2B ) ., However , Naip2 did not associate with Nlrc4 in uninfected macrophages or macrophages infected with the mutant bacterium lacking a functional T3SS that are unable to release MxiI into the host cytosol ( Figure 2B ) ., Furthermore , infection with Shigella preferentially promoted the interaction of Nlrc4 with Naip2 relative to Naip5 ( Figure 2B ) ., MxiI is secreted into the culture medium by Shigella which relies on the presence of a functional T3SS 19–21 ., Therefore , MxiI is presumably leaked into the host cytosol via the T3SS to activate Nlrc4 , as it was suggested for Salmonella PrgJ 22 , 23 ., Therefore , we next asked whether expression of MxiI promotes the association of Naip2 with endogenous Nlrc4 in uninfected macrophages ., Immunoprecipitation experiments showed that expression of MxiI induced the interaction of Naip2 with endogenous Nlrc4 ( Figure 2C ) ., Collectively , these results indicate that MxiI interacts preferentially with Naip2 and promotes the interaction between Naip2 and Nlrc4 ., We next performed additional studies to verify that Shigella infection promoted the activation of Nlrc4 via Naip2 ., To confirm the preferential effect of Naip2 on Nlrc4 activation , we performed reconstitution experiments by expressing Nlrc4 , Asc , caspase-1 , pro-IL-1β and Naip2 or Naip5 in 293T cells ., One day after transfection , cells were infected with WT or T3SS-deficient Shigella for 3 hrs and inflammasome activation was analyzed by immunoblotting with an antibody specific for mature IL-1β p17 ., In the absence of exogenous Naip2 or Naip5 , infection with WT Shigella enhanced the processing of pro-IL-1β into IL-1β p17 ( Figure S1A ) ., The formation of IL-1β p17 was further enhanced by Naip2 , but inhibited by Naip5 in Shigella-infected cells ( Figure S1A ) ., In this reconstitution system , the enhancement of IL-1β p17 formation by Naip2 in cells infected with WT Shigella required Nlrc4 , Asc and caspase-1 ( Figure S1B ) ., Shigella infection stimulates Nlrc4- and Asc-dependent inflammasome activation in macrophages 18 ., However , Shigella was also shown to induce macrophage cell death via Nlrp3 after 2–6 hrs of infection at a bacteria/macrophage ratio of 50∶1 24 ., To verify these seemingly contradictory results , we reassessed the role of Asc , Nlrc4 and Nlrp3 in Shigella-induced caspase-1 activation ., In these experiments , LPS-primed BMDM were infected with the Shigella WT or S325 ( T3SS-deficient mutant ) at a bacteria/macrophage ratio of 10∶1 for 30 min ., As expected , WT , but not mutant Shigella , induced processing of procaspase-1 into the p20 subunit of caspase-1 ( Figure S2A ) ., The inability of the mutant bacterium to activate caspase-1 could not be explained by reduced uptake by macrophages ( Figure S3 ) ., Importantly , caspase-1 activation , IL-1β release , and pyroptosis required Nlrc4 and Asc , but not Nlrp3 ( Figure S2A–C ) ., Because previous studies showed that Asc was not required for pyroptosis induced by Shigella in BMDM differentiated for 5 days 18 , we assessed cell death induced by Shigella in BMDM differentiated for 3 , 4 and 5 days in culture ( Figure S4 ) ., Consistent with previous studies 18 , Asc was not required for pyroptosis in macrophages differentiated for 5 days ( Figure S4 ) ., In macrophages differentiated for 3 or 4 days , however , cell death induced by Shigella was enhanced in WT macrophages and impaired in Asc-deficient macrophages ( Figure S2C and S4 ) which is in line with the results presented in Figure S2C ., Next , we investigated the role of Naip2 and Naip5 in caspase-1 activation induced by Shigella ., We used siRNA-mediated knockdown to reduce the expression of Naip2 and Naip5 in macrophages ( Figure 3A ) ., Notably , caspase-1 activation induced by Shigella was attenuated by inhibiting the expression of Naip2 , but not Naip5 ( Figure 3B ) ., Importantly , the ability of individual siRNA to inhibit caspase-1 activation correlated with reduction of Naip2 expression ( Figure 3A , B ) ., In addition , knockdown of Naip2 , but not Naip5 , reduced the release of IL-1β and IL-18 induced by Shigella infection at 1 or 2 hrs post-infection ( Figure 3B , C ) ., In control experiments , knockdown of Naip2 did not affect the production of IL-6 or CXCL2 in macrophages infected with WT or S325 mutant Shigella ( Figure 3C ) ., These results suggest that Shigella induces Nlrc4-dependent inflammasome activation via Naip2 in macrophages ., The Asc pyroptosome is a molecular platform that is thought to be important for the recruitment and activation of caspase-1 25–27 ., Infection of macrophages with WT , but not T3SS-deficient , Shigella induced the formation of the Asc pyroptosome which was detected in the cell cytoplasm by staining with an antibody that recognizes Asc ( Figure 4A , B ) ., The Asc pyroptosome induced by Shigella infection co-localized with FLICA staining that labels activated caspase-1 ( Figure 4A ) ., Importantly , knockdown of Naip2 by siRNA reduced Asc pyroptosome formation whereas Naip5 did not ( Figure 4C , D ) ., To provide direct biochemical evidence that the Asc pyroptosome is formed , we cross-linked the insoluble Asc protein complexes from Shigella or Salmonella infected macrophages and subjected them to immunoblotting with anti-Asc antibody ., Immunoblotting analysis revealed that infection with WT Shigella or Salmonella induces prominent Asc dimer formation in WT , but not Asc-deficient macrophages ( Figure 5A , upper panel ) ., The induction of Asc dimers correlated with IL-1β release in culture supernatants ( Figure 5A , lower panel ) ., In contrast , Shigella deficient in T3SS and the fliA-deficient Salmonella mutant were impaired in the induction of Asc dimer formation ( Figure 5A ) ., Notably , expression of MxiI was sufficient to induce the formation of Asc dimers in caspase-1-deficient macrophages in the absence of Shigella infection ( Figure 5B ) ., Furthermore , knockdown of Naip2 by siRNA , but not Naip5 , inhibited Asc dimer formation ( Figure 5C ) ., These results indicate that Shigella MxiI and Naip2 are important in Asc pyroptosome formation which is associated with inflammasome activation ., Recent studies reported that Nlrc4 phosphorylation by Pkcδ is critical for inflammasome activation induced by Salmonella infection 13 ., Thus , we assessed whether inflammasome activation caused by Shigella infection also requires Pkcδ ., In these experiments , LPS-primed BMDM from WT and Pkcδ-deficient mice were infected with WT or S325 ( T3SS-deficient mutant ) Shigella , and IL-1β release was evaluated at different time points and bacterial/macrophage ratios after infection ., As expected , expression of Pkcδ was induced by LPS stimulation in WT , but not Pkcδ-deficient macrophages ( Figure 6A ) ., Importantly , Pkcδ was not required for IL-1β secretion induced by Shigella or Salmonella ( Figure 6B–D ) ., In fact , Pkcδ deficiency enhanced IL-1β secretion in response to Shigella and Salmonella infection ( Figure 6B–D ) ., Furthermore , Pkcδ-deficient macrophages produced higher amounts of IL-1α and CCL5 , but not CXCL2 than WT macrophages in response to infection ( Figure 6D ) ., The increased production of cytokines in Pkcδ-deficient macrophages was not associated with enhanced NF-κB or MAPK activation after Shigella infection ( Figure S5 ) ., Notably , induction of apoptosis in Shigella-infected macrophages was inhibited in macrophages deficient in Pkcδ ( Figure S5 ) ., Furthermore , treatment with z-DEVD-fmk , a cell permeable caspase-3 inhibitor , increased the production of IL-1β in WT macrophages infected with Shigella ( Figure S5 ) , suggesting that increased production of IL-1β in Pkcδ-deficient macrophages is mediated , at least in part , by inhibition of apoptosis in Shigella-infected macrophages ., Importantly , caspase-1 activation induced by Shigella or Salmonella was unimpaired in macrophages deficient in Pkcδ ( Figure 6E ) , whereas it was abolished in macrophages deficient in Nlrc4 ( Figure 6E ) ., These results indicate that Pkcδ is not essential for inflammasome activation induced by Shigella or Salmonella infection ., The intracellular sensing of flagellin is the major trigger for the activation of the Nlrc4 inflammasome in macrophages infected with Salmonella 4 ., Because Shigella is non-flagellated , the current studies were aimed at understanding the mechanism by which Shigella induces the activation of Nlrc4 in macrophages ., We show here that Shigella induces the activation of the Nlrc4 inflammasome through MxiI , an inner rod protein of the T3SS ., MxiI associated with Naip2 and was sufficient to induce Nlrc4-dependent IL-1β secretion and the interaction with Nlrc4 ., Importantly , inhibition of Naip2 expression impaired the activation of the Nlrc4 inflammasome and IL-1β/IL-18 release in Shigella-infected macrophages ., Because IL-1β secretion induced by Shigella was not abolished by Naip2 knockdown , it is possible that Shigella also activates another inflammasome pathway that is minor and only unmasked by the inhibition of the Naip2-Nlrc4 pathway ., Alternatively , it is possible that the partial inhibition of IL-1β secretion reflects residual Naip2 protein expression in macrophages ., Our work is consistent with a model in which the T3SS inner rod proteins including PrgJ in Salmonella and MxiI in Shigella are recognized by Naip2 and this interaction leads to the recruitment and activation of Nlrc4 ., Consistent with this model , we show that expression of MxiI promotes the association of Naip2 with Nlrc4 and induces the oligomerization of Asc in macrophages ., Furthermore , WT , but not T3SS-deficient Shigella , enhances the association of Naip2 and Nlrc4 in macrophages ., The failure of mutant Shigella to induce the interaction between Naip2 and Nlrc4 is presumably explained by the inability of the T3SS mutant to release MxiI into the host cytosol ., A measure of inflammasome activation is the formation of Asc oligomers 25–27 ., Importantly , Asc oligomerization induced by MxiI was observed in caspase-1-deficient macrophages , indicating that this critical event is not a secondary event of caspase-1 activation ., MxiI is composed of 97 amino acids and is predicted to be a soluble protein using publically available tools ( http://www . psort . org/psortb ) ., It has been shown that MxiI is secreted into the culture medium by Shigella in a T3SS dependent manner 19–21 ., Thus , as it was suggested for Salmonella PrgJ 9 , 22 , we propose that small amounts of MxiI are leaked into the host cytosol via the T3SS during Shigella infection to induce the activation of Nlcr4 ., Recent studies showed that Nlrc4 phosphorylation was induced by Salmonella and was found to be critical for inflammasome activation 13 ., Furthermore , it was proposed that Pkcδ was the major kinase responsible for phosphorylation of Nlrc4 13 ., In contrast to the latter finding , we found that IL-β secretion and caspase-1 activation induced by Shigella and Salmonella infection were not impaired in Pkcδ-deficient macrophages ., Notably , the production of several cytokines including IL-1β was enhanced in infected Pkcδ-deficient macrophages ., A possible mechanism to account for the enhanced production of cytokines in Pkcδ-deficient macrophages is the observation that Pkcδ regulates phagosomal production of ROS 28 which is known to inhibit pro-inflammatory responses including cytokine production 29 ., However , we did not observe enhanced NF-κB or MAPK activation in Pkcδ-deficient macrophages infected with Shigella ., Pkcδ has been shown to regulate the induction of apoptosis 30–32 ., Consistently , apoptosis induced by Shigella infection was impaired in Pkcδ-deficient macrophages and treatment with a caspase-3 inhibitor enhanced IL-1β secretion in WT macrophages ., These results suggest that the increased production of cytokines observed in Pkcδ-deficient macrophages might be due , at least in part , to suppression of apoptosis in infected macrophages ., Regardless of the mechanism involved , our results clearly show that caspase-1 activation induced by Shigella or Salmonella infection is not impaired in Pkcδ-deficient macrophages ., We do not have a clear explanation for the difference in results between our studies and previous results by Qu et al . These authors showed that in addition to Pkcδ , Pak2 was capable of phosphorylating Nlrc4 at the critical Ser533 , although the results suggested that Pak2 was a minor Nlrc4-phosphorylating kinase 13 ., Thus , it is conceivable that the difference in results could be explained by kinase redundancy and subtle variation in the expression of Nlrc4-phosphorylating kinases in different macrophage preparations ., Regardless of the explanation , findings within this investigation clearly show that Pkcδ is dispensable for Nlrc4 activation ., Thus , our results challenge the notion that Pkcδ is critical for inflammasome activation and indicate that further work is needed to understand the mechanism and role of Nlrc4 phosphorylation in inflammasome activation ., Shigella MxiI associates with Naip2 to induce the interaction of Naip2 with Nlrc4 , which presumably leads to Nlrc4 oligomerization and inflammasome activation ., In the Salmonella system , cytosolic flagellin binds to Naip5 and induces the association of Naip5 with Nlrc4 10–12 ., Reconstitution experiments with purified flagellin , Naip5 and Nlrc4 revealed that these components are sufficient to induce the formation of a disk-like complex composed of 11 or 12 proteins including Nlrc4 and Naip5 , although the exact ratio of Naip5 and Nlrc4 in the complex remains unclear 12 ., Based on the latter observations , we suggest that Shigella MxiI induces the oligomerization of Nlrc4 via their interaction with Naip2 ., Consistent with this model , we found that MxiI induced the interaction of Naip2 with Nlrc4 and the oligomerization of Asc ., Furthermore , Naip2 , but not Naip5 , was critical for caspase-1 activation , pyroptosome formation , Asc oligomerization and IL-1β secretion ., Collectively , these results support a model in which distinct Naip family members act as sensors of flagellin and T3SS inner rod proteins and oligomerized Nlrc4 provides a platform for the recruitment and activation of caspase-1 ., While Naip2 knockdown reduced inflammasome activation , Naip5 knockdown had the opposite effect in response to Shigella infection ., Although further work is needed to understand the role of Naip5 , one possibility is that there is competition between Naip2 and Naip5 protein complexes and inhibition of Naip5 enhances the Naip2-Nlrc4 inflammasome pathway ., Nlrc4 and caspase-1 contain CARD domains and they could interact directly via homotypic CARD-CARD interactions ., However , the adaptor Asc is essential for the activation of caspase-1 in response to Salmonella and Shigella 18 , 33 ., These results suggest that Asc is somehow required for the interaction between Nlrc4 and caspase-1 or that Asc is critical for another step which is important for inflammasome activation ., All animal experiments were conducted according to the U . S . A . Public Health Service Policy on Humane Care and Use of Laboratory Animals ., Animals were maintained in an AAALAC approved facility and all animal studies followed protocol 09716-2 that was approved by the Animal Care and Use Committee of the University of Michigan ( Ann Arbor , MI ) ., Mice deficient in Nlrc4 , Nlrp3 , Asc and caspase-1/11 have been previously described 4 , 34 , 35 ., All mice were crossed at least 5 times on a C57BL/6 background ., Bone marrow samples from Prkcd−/− mice in C57BL/6 background were provided by Hee-Jeong Im Sampen ( Rush University Medical Center , Chicago , IL ) ., Shigella flexneri strain YSH6000 36 was used as the WT strain , and S325 ( mxiA::Tn5 ) 37 was used as the T3SS–deficient control ., The WT S . enterica serovar Typhimurium SR-11 χ3181 and the isogenic fliA::Tn10 were provided by H . Matsui ( Kitasato Institute for Life Science , Tokyo , Japan ) 38 ., ΔfliA Salmonella mutant is impaired in the expression of flagellin 18 ., cDNAs encoding mouse Naip2 , Naip5 , Nlrc4 , Asc , caspase-1 , and bacterial MxiI were amplified by PCR and cloned into the pCMV based mammalian expression vector or the MSCV-IRES-GFP retroviral expression vector ( Addgene ) ., Human pro-IL-1β clone ( RDB6666 ) was provided by RIKEN BRC which is participating in the National Bio-Resource Project of the MEXT , Japan ., BMDMs were prepared from the femurs and tibias of mice and cultured for 3–7 days in 10% FCS IMDM ( Gibco ) supplemented with 30% L-cell supernatant , non-essential amino acids , sodium pyruvate and antibiotics ( Penicillin/Streptomycin ) ., 293T cells were cultured on Dulbeccos Modified Eagles medium ( Sigma ) containing 10% FCS and antibiotics ( Penicillin/Streptomycin ) ., The rabbit anti mouse caspase-1 p20 and anti-mouse Nlrc4 antibodies were produced in our laboratory by immunizing rabbits with mouse caspase-1 ( p20 subunit ) and mouse Nlrc4 ( amino acids 1–152 ) recombinant proteins 39 ., Anti–IL-1β p17 ( #2021 ) and anti-Pkcδ ( #2058 ) antibodies were from Cell Signaling ., Mouse monoclonal anti-β-actin antibody was from Sigma ., HRP-conjugated goat anti–rabbit ( Jackson Laboratories ) or anti–mouse IgG ( Sigma ) or anti-rat ( Jackson Laboratories ) , or AP-conjugated goat anti-rabbit ( Santa Cruz Biotechnology Inc . ) or anti-mouse IgG ( Santa Cruz Biotechnology Inc . ) antibodies were used as secondary antibodies for immunoblotting ., Macrophages were seeded in 24-well plates at a density of 3×105 cells per well ., Cells were stimulated with or without 0 . 1 µg/ml LPS ( from E . coli O55:B5 , Sigma ) for 6 h and then infected with Shigella or Salmonella ., Bacterial strains were pre-cultured overnight in Mueller-Hinton broth ( Difco ) at 30°C , then were inoculated into brain heart infusion broth ( Difco ) and incubated for 2 h at 37°C prior to infection ., The cells were infected with Shigella at a bacteria/macrophage ratio of 10∶1 , or with Salmonella at a bacteria/macrophage ratio of 1∶1 unless otherwise stated ., The plates were centrifuged at 700 g for 5 min to synchronize the infection , and gentamicin ( 100 µg/ml ) and kanamycin ( 60 µg/ml ) were added after 20 min ., At the indicated times after infection , cytokines were measured in culture supernatants by enzyme-linked immunoabsorbent assay ( ELISA ) kits ( R and D Systems ) ., RNA was isolated with E . Z . N . A . TM total RNA kit ( Omega Biotek ) according to the manufacturers instructions ., RNA was reverse transcribed using the High Capacity RNA-to cDNA kit ( Applied Biosystem ) and cDNA was then used for RT-PCR ., For immunofluorescence studies , the infected cells were fixed and immunostained , and then analyzed with a confocal laser-scanning microscope ( LSM510; Carl Zeiss ) or fluorescence microscopy ( Olympus ) ., Carboxyfluorescein FLICA ( Immunochemistry Technologies , LLC ) was added 1 hr before bacterial infection ., Apoptosis was measured by the AnnexinV ( Roche ) and TUNEL ( Promega ) assays using fluorometric protocols according to the manufactures recommendations ., For the caspase-3 inhibitor studies , the cells were treated with 200 µM z-DEVD-fmk ( Calbiochem ) for 1 h before bacterial infection ., 293T cells were seeded in 6-well plates at a density of 5×105 cells per well and incubated overnight ., Then , the cells were transfected with or without 1 µg T7-tagged Nlrc4 , 1 µg T7-tagged Asc , 0 . 4 µg HA-tagged caspase-1 , and 0 . 4 µg FLAG-tagged proIL-1β 40 , and 1 µg HA-tagged Naip2 or Naip5 , using FuGENE 6 ( Roche ) ., Cells were infected one day after infection ., Intensities of casp1 p20 or IL-1β p17 bands were quantified by densitometry , the values normalized to the β-actin protein levels and results were analyzed with ImageJ software ., The Shigella MxiI gene was cloned into the MSCV-IRES-GFP retrovirus vector , which contains an IRES-GFP element to track retroviral infection ., WT or Nlrc4−/− BMDMs were immortalized using the J2 virus to increase nucleofection efficiency 41 ., Then , cells were nucleofected with MSCV-IRES-GFP or MSCV-IRES-GFP encoding Shigella MxiI using an Amaxa nucleofector system ( Nucleofector kit V and the D-032 program ) ., After 20 hrs , cell survival in the GFP-positive cell population was analyzed by fluorescence microscopy ., The LDH activity in the culture supernatants of infected cells was measured using the CytoTox 96 assay kit ( Promega ) according to the manufacturers protocol ., Assays were performed in triplicate for each independent experiment ., The invasion efficiency of Shigella strains was evaluated using a gentamicin/kanamycin protection assay ., Briefly , cells were infected for 20 min and then incubated for 20 min at 37°C in medium containing gentamicin ( 100 µg/ml ) and kanamycin ( 60 µg/ml ) to kill extracellular bacteria ., The infected cells were then washed in PBS , lysed in 0 . 5% TritonX-100/PBS , and serial dilutions were plated on LB agar plates to determine the number of intracellular bacteria ., DNA and siRNAs specific for Naip2 and Naip5 were transfected into macrophages using an Amaxa nucleofector system ( Y-001 program for primary macrophages or D-032 program for cell lines ) according to the manufacturers instructions ., siRNA pools for mouse Naip2 ( 17948; D-044151-01-04 ) and Naip5 ( 17951; D-044141-01-4 ) and non-targeting siRNAs were purchased from Dharmacon or synthesized by Sigma and targeting the sequences CTTACACTGAATCACAAGA ( naip2 ) or GTGCCTTTTTAGTCCTTGT ( naip5 ) ., Primer sets for RT-PCR were naip2-forward ( AGGCTATGAGCATCTACCACA ) , naip2-reverse ( AAGACATCAATCCACAGCAAA ) , naip5-forward ( TGCCAAACCTACAAGAGCTGA ) , naip5-reverse ( CAAGCGTTTAGACTGGGGATG ) , actin-forward ( CATGTACGTTGCTATCCAGGC ) and actin-reverse ( CTCCTTAATGTCACGCACGAT ) ., To compare caspase-1 p20 levels in immunoblotting experiments , the bands were quantified by densitometry , analyzed with ImageJ software , and normalized to the β-actin protein levels ., Cell ware lysed in IP buffer CelLytic M Cell Lysis Reagent ( Sigma ) , 0 . 1 mM PMSF , and a complete protease inhibitor cocktail-EDTA ( Roche ) and clarified lysates were mixed with anti-T7 antibody–conjugated agarose beads ( Novagen ) or anti-HA conjugated sepharose beads ( Covance ) for 1 hr at 4°C with gentle rotation in IP buffer . Beads were washed with PBS , mixed with SDS-sample buffer and subjected to immunoblot analysis . Cells were fixed with 4% paraformaldehyde and 0 . 1% NP40 , washed and stained with anti- Asc antibody and FITC-conjugated anti–rat antibody ( Sigma ) as a secondary antibody . Imaging analysis was performed using fluorescence microscopy ( Olympus ) , and percentage of cells containing Asc pyroptosomes was determined by counting at least 300 cells in 5 separate fields . The Asc dimerization assay was previously described 25–27 ., Briefly , cells were lysed ( 20 mM HEPES-KOH , pH 7 . 5 , 150 mM KCl , 1% NP-40 , 0 . 1 mM PMSF , and Complete protease inhibitor cocktail ( Roche ) ) and forced onto a 21-gauge needle 10 times ., The cell lysates were centrifuged at 6000 rpm for 10 min at 4°C to isolate the insoluble fraction in the pellet ., The pellets were washed twice with PBS , resuspended in 500 µl of PBS and cross-linked with fresh 2 mM disuccinimidyl suberate ( DTT , Sigma ) for 30 min ., The cross-linked pellets were isolated by centrifugation at 13000 rpm for 10 min and resuspended in 20 µl of SDS sample buffer for immunoblotting with anti-mouse Asc antibody ., Mouse cytokines in culture supernatants were measured by ELISA kits ( R&D Systems ) ., Assays were performed in triplicate for each independent experiment ., Statistical analyses were performed using the Mann–Whitney U test ., Differences were considered significant at p<0 . 05 .

Introduction, Results, Discussion, Materials and Methods

Recognition of intracellular pathogenic bacteria by members of the nucleotide-binding domain and leucine-rich repeat containing ( NLR ) family triggers immune responses against bacterial infection ., A major response induced by several Gram-negative bacteria is the activation of caspase-1 via the Nlrc4 inflammasome ., Upon activation , caspase-1 regulates the processing of proIL-1β and proIL-18 leading to the release of mature IL-1β and IL-18 , and induction of pyroptosis ., The activation of the Nlrc4 inflammasome requires the presence of an intact type III or IV secretion system that mediates the translocation of small amounts of flagellin or PrgJ-like rod proteins into the host cytosol to induce Nlrc4 activation ., Using the Salmonella system , it was shown that Naip2 and Naip5 link flagellin and the rod protein PrgJ , respectively , to Nlrc4 ., Furthermore , phosphorylation of Nlrc4 at Ser533 by Pkcδ was found to be critical for the activation of the Nlrc4 inflammasome ., Here , we show that Naip2 recognizes the Shigella T3SS inner rod protein MxiI and induces Nlrc4 inflammasome activation ., The expression of MxiI in primary macrophages was sufficient to induce pyroptosis and IL-1β release , which were prevented in macrophages deficient in Nlrc4 ., In the presence of MxiI or Shigella infection , MxiI associated with Naip2 , and Naip2 interacted with Nlrc4 ., siRNA-mediated knockdown of Naip2 , but not Naip5 , inhibited Shigella-induced caspase-1 activation , IL-1β maturation and Asc pyroptosome formation ., Notably , the Pkcδ kinase was dispensable for caspase-1 activation and secretion of IL-1β induced by Shigella or Salmonella infection ., These results indicate that activation of caspase-1 by Shigella is triggered by the rod protein MxiI that interacts with Naip2 to induce activation of the Nlrc4 inflammasome independently of the Pkcδ kinase .

Shigella are bacterial pathogens that are the cause of bacillary dysentery ., An important feature of Shigella is their ability to invade the cytoplasm of host epithelial cells and macrophages ., A major component of host recognition of Shigella invasion is the activation of the inflammasome , a molecular platform that drives the activation of caspase-1 in macrophages ., Although Shigella is known to induce the activation of the Nlrc4 inflammasome , the mechanism by which the bacterium activates Nlrc4 is largely unknown ., We discovered that the Shigella T3SS inner rod protein MxiI induces Nlrc4 inflammasome activation through the interaction with host Naip2 , which promoted the association of Naip2 with Nlrc4 in macrophages ., Expression of MxiI induced caspase-1 activation , Asc oligomerization , pyroptosis and IL-1β release which required Naip2 , but not Naip5 ., Significantly , caspase-1 activation induced by Shigella infection was unaffected by deficiency of the Pkcδ kinase ., This study elucidates the microbial-host interactions that drive the activation of the Nlrc4 inflammasome in Shigella-infected macrophages .

medicine, bacterial diseases, infectious diseases, immunity, innate immunity, immunology, biology

journal.ppat.1004953

Transmission Properties of Human PrP 102L Prions Challenge the Relevance of Mouse Models of GSS

Prion diseases are a closely related group of neurodegenerative conditions which affect both humans and animals 1 , 2 ., They are both experimentally and , in some cases , naturally transmissible within and between mammalian species ., Cross-species transmission is generally much less efficient than within-species transmissions , being limited by a ‘species’ or transmission barrier 2 , 3 ., Prion diseases in humans include Creutzfeldt-Jakob disease ( CJD ) , Gerstmann-Sträussler-Scheinker disease ( GSS ) , fatal familial insomnia ( FFI ) , kuru and variant CJD ( vCJD ) 1 , 4 , 5 ., According to the widely accepted ‘protein-only’ hypothesis 6 , the central feature of prion disease is the conversion of host-encoded cellular prion protein ( PrPC ) to alternative isoforms designated PrPSc 1 , 2 , 7 ., It is proposed that PrPSc is the infectious agent acting to replicate itself with high fidelity by recruiting endogenous PrPC and that the difference between these isoforms lies purely in the monomer conformation and its state of aggregation 1 , 2 , 8 although it is now clear that infectivity can also be associated with protease-sensitive disease-related PrP assemblies distinct from classical PrPSc 9–11 and that infectious and neurotoxic PrP species can be uncoupled 12 , 13 ., Inherited prion disease ( IPD ) is caused by autosomal-dominant mutations in the human PrP gene ( PRNP ) and constitute about 15% of all human prion disease 4 , 14 ., Over 40 mutations have been identified , but the precise biochemical mechanisms that lead to disease remain unknown ., Within the framework of the protein-only hypothesis , pathogenic mutations in PrP are thought to predispose the mutant proteins to adopt disease-causing conformations and assembly states 2–4 ., A proline to leucine substitution at codon 102 ( P102L ) of human PrP is the most common mutation associated with the GSS phenotype and was first reported in 1989 15 ., Many other kindreds have now been documented worldwide 16 , including the original Austrian family reported by Gerstmann , Sträussler , and Scheinker in 1936 17 , 18 ., Progressive ataxia is the dominant clinical feature , with dementia and pyramidal features occurring later in a disease course typically much longer than that of classical CJD ., However , marked variability at both the clinical and neuropathological levels is apparent , with some patients developing a classical CJD-like phenotype with early and rapidly progressive dementia 18–30 ., A significant part of this phenotypic variability appears to be contributed by variable propagation of distinct disease-related PrP species generated from either PrP 102L 21 , 22 or wild type PrP 24 , 27 ., Two distinct abnormal conformers of PrP 102L that generate proteinase K ( PK ) -resistant fragments of either ~21–30 kDa or ~8 kDa 21 , 22 , 24 , 27 , 31 have distinct prion transmission properties in 101LL PrP gene knock-in mice 25 , while the potential transmissibility or neurotoxicity of abnormal conformers of wild type PrP ( that generate PK-resistant fragments of 21–30 kDa 24 , 27 ) remains unknown ., Such heterogeneity in disease-related PrP isoforms present in IPD P102L patient brain severely complicates interpretation of transmissions in both conventional and transgenic mice ., The conformational selection hypothesis 2 , 32 predicts that heterogeneous prions formed from PrP in distinct conformations would be differentially selected by hosts expressing different PrP primary sequences ., In this regard expression of the homotypic human mutant protein in the host may be critical to accurately model the disease , as only the human mutant protein may be conformationally susceptible to the prion strain involved 2 , 3 , 33 ., Much of the transgenic modelling of inherited prion disease has however focused on superimposing human PrP mutations onto rodent PrP in order to establish whether infectious prions can be generated de novo ., An extremely important consideration in such studies is whether superimposition of pathogenic human PrP mutation into mouse PrP will have the same structural consequences 2 , 3 , 33 , 34 ., The possibility of propagating novel prion strains that do not recapitulate the molecular and neuropathological phenotype of the original human disease appears probable 2 , 3 , 33 and indeed has been documented with variant CJD transmissions 35 ., Recently we established that IPD P102L patient brain isolates could transmit disease with 100% clinical attack rates and short incubation periods to transgenic mice expressing human PrP 102L on a mouse PrP null background ( designated 102LL Tg27 mice ) 33 ., In these transmissions we observed the propagation of the abnormal conformer of PrP 102L that generates protease-resistant fragments of ~21–30 kDa 33 ., We also demonstrated that such mice were susceptible to infection with classical CJD prions leading to the generation of prions with altered PrPSc glycoform ratios 33 ., The availability of these prions from 102LL Tg27 mice , in which disease-related PrP is entirely composed of PrP 102L ( as opposed to the heterogeneous PrP in primary human GSS brain inoculum ) , now permits direct testing of their host range and in particular the ability of these prions to propagate using wild type human PrP or mouse PrP as substrate ., Our findings show that human PrP 102L can support the propagation of distinct prion strains and that human PrP 102L prions have transmission properties strikingly different from those generated in transmission models in which the 102L mutation was superimposed onto mouse PrP ., Prions originating from the primary transmission of three different IPD P102L patient brains to 102LL Tg27 mice 33 ( hereafter designated GSS-102L prions ) transmitted clinical prion disease with 100% attack rates and short mean incubation periods ( ~165 days ) when passaged in further 102LL Tg27 mice ( Table 1 ) ., Brain samples of all mice in these transmissions were positive for PK-resistant PrP 102L by immunoblotting using ICSM 18 ( Fig 1A ) ( Table 1 ) ., Both the PK-resistant PrP fragment size ( ~21–30 kDa ) and predominance of the di-glycosylated PrP glycoform mirrored that seen in the 102LL Tg27 mouse brain inoculum ( Fig 1A ) ., Immunohistochemistry with ICSM 18 showed extensive abnormal PrP deposition throughout the brain ( thalamus shown in Fig 1C ) accompanied by prominent astrocytosis and spongiosis ., Collectively , these findings establish that IPD P102L prions propagate with high efficiency when serially passaged in 102LL Tg27 mice ., Much of the previous modeling of IPD P102L has involved superimposing the mutation onto the wild type mouse PrP sequence ( reviewed in ref 3 ) ., However , it is unclear if challenge of mouse PrP 101L with human P102L prions would lead to the generation of authentic human prion strains or conversely would lead to the generation of experimental prion strains with different transmission characteristics ., Notably , after human P102L prions were passaged once in 101LL PrP gene knock-in mice the resultant prions were shown to readily infect wild type mice 36 ., We therefore inoculated the GSS-102L prion isolates that transmitted efficiently on passage in 102LL Tg27 mice to wild type mice and also to transgenic mice expressing wild type human PrP on a mouse PrP null background ., Strikingly , in complete contrast to findings with the mouse 101LL PrP gene knock-in model we found that GSS-102L prions failed to produce clinical prion disease or any evidence of sub-clinical prion infection when inoculated into wild type mice ( Table 1 ) ., Even more remarkably the same GSS-102L prions produced no clinical prion disease or evidence of sub-clinical prion infection when inoculated into transgenic mice expressing wild type human PrP ( Table 1 ) ., In all of these negative transmissions examination of brain showed no detectable PrPSc by high sensitivity immunoblotting ( Fig 1A and 1B , lanes 4 and 5 ) or abnormal PrP deposition by immunohistochemistry ( Fig 1E–1H ) ., In addition , no evidence for elevated levels of spongiosis or gliosis in comparison to the brain of uninoculated age-matched control mice was observed ., Collectively these data establish that GSS-102L prions which replicate with high efficiency in a host expressing human PrP 102L are unable to propagate using wild type human PrP or wild type mouse PrP as substrate ., In comparison to IPD P102L prions , transmission of classical CJD prions to 102LL Tg27 mice appears to be limited by a transmission barrier 33 ., In primary transmissions , although nearly all CJD prion-challenged 102LL Tg27 mice showed evidence for prion infection , only a proportion of mice developed clinical prion disease and then only after prolonged incubation periods 33 ., In addition , a change in propagated PrPSc type was observed ( which itself is indicative of a transmission barrier 2 ) with PrPSc glycoform ratios switching from those present in the CJD inocula to ones that more closely resemble those seen in the brain of IPD P102L patients and IPD P102L prion-challenged Tg27 mice 33 ., From these primary transmissions , we were unable to distinguish whether the 102L mutation in the host PrP had directly dictated the strain characteristics of the propagated prions ( to essentially become congruent with GSS-102L prions ) or whether CJD-like prion strain properties were retained ., To investigate this , we passaged prions from CJD-challenged 102LL Tg27 mice ( hereafter designated CJD-102L prions ) in further 102LL Tg27 mice , in transgenic mice expressing wild type human PrP and in wild type mice ( Table 2 ) ., In 102LL Tg27 mice we observed that the barrier to development of clinical prion disease seen at primary transmission of classical CJD prions was not abrogated at secondary passage ( Table 2 ) ., Although nearly all CJD-102L prion-inoculated mice developed prion infection , as evidenced by detection of PrPSc ( Fig 2 ) and abnormal PrP deposition throughout the brain ( Fig 3 ) , clinical prion disease was again only observed in a proportion of inoculated recipients and then only at prolonged incubation periods ( Table 2 ) ., PrPSc typing showed that the altered PrPSc glycoform ratio of CJD-102L prions generated after primary transmission of classical CJD prions to 102LL Tg27 mice was not maintained after further passage in the same mice ., The PrPSc type now appeared to more closely resemble the original classical CJD inoculum with a predominance of mono-glycosylated PrP and was readily distinguishable from the di-glycoslyated PrP dominant glycoform pattern seen after secondary passage of GSS 102L prions in 102LL Tg27 mice ( Fig 2A , 2C and 2E ) ( Table 3 ) ., From these transmissions we concluded that GSS-102L prions and CJD-102L prions have incongruent transmission properties after further passage in 102LL Tg27 mice ., Importantly , the disparate nature of CJD-102L prions and GSS-102L prions became obvious after examining the transmission properties of CJD-102L prions in transgenic mice expressing wild type human PrP ., In complete contrast to GSS-102L prions , all CJD-102L prion isolates transmitted clinical prion disease to mice expressing wild type human PrP in a fashion analogous to the original CJD inoculum ( Table 2 ) ., In these transmissions PrPSc was readily detected in brain by immunoblotting ( Fig 2 ) and abnormal PrP deposition was observed throughout the brain by immunohistochemistry ( Fig 3 ) ., Humanised transgenic mice expressing human PrP 129 valine on a Prnp null background are highly susceptible to sporadic CJD prions regardless of the PrPSc type or codon 129 genotype of the inoculum 37–43 ., These transmissions are typically characterised by 100% attack rates of prion infection producing uniform clinical prion disease after similar short incubation periods of around 200 days ., The absence of a transmission barrier to sporadic CJD prions is not however uniformly observed in transgenic mice expressing human PrP 129 methionine on a Prnp null background ., Here mismatch at residue 129 between the inoculum and host can significantly affect transmission 41 , 44–46 as evidenced by more prolonged and variable incubation periods and reduced attack rates 41 , 43 , 44 ., Remarkably , we observed that CJD-102L prions behaved in a closely similar fashion that corresponded with the codon 129 status of the original CJD inoculum ( Table 2 ) ., This was striking because all of the CJD-102L prion isolates have PrP with residue 129 methionine ., Consistent with the CJD-like transmission properties of CJD-102L prions in transgenic mice expressing wild type human PrP , PrPSc typing of the recipient mouse brain showed that the di-glycosylated dominant PrPSc glycoform ratio of CJD-102L prions in the inoculum had switched to a mono-glycosylated PrPSc dominant pattern which more closely resemble CJD prions ( Table 3; Fig 2B , 2D and 2F , lanes 5 ) ., Collectively , these data show that CJD-102L prions are distinct from GSS-102L prions and retain the transmission properties of the original CJD prion strains ., Notwithstanding these observations , all the CJD-102L prion isolates were obtained after a single passage of classical CJD prions in 102LL Tg27 mice and it remains to be seen whether serial passage on the mutated sequence would lead to similar conservation of CJD phenotype ., Consistent with the finding that classical CJD prions transmit prion infection only occasionally to wild type mice with long and variable incubation periods 37 , 39 , 40 , 42 , 47 we found that CJD-102L prions were also unable to propagate efficiently in wild type mice ( Table 2 ) ., We found that only one out of eighteen CJD-102L prion-inoculated wild type mice became infected ( Table 2 ) with all other mice showing no evidence of subclinical prion infection by either PrP immunoblotting or immunohistochemistry ., Co-propagation of distinct disease-related PrP conformers in IPD brain , combined with differences in their neuropathological targeting , abundance and potential neurotoxicity , provides a general molecular mechanism underlying phenotypic heterogeneity in patients with the same PRNP mutation ., Previously we and others have reported the propagation of distinct isoforms of protease-resistant PrP with divergent properties in IPD P102L patient brain and such molecular heterogeneity severely hampers interpretation of primary transmissions to both conventional and transgenic mice ( for review see 3 ) ., In the present study we have investigated the properties of prions generated in transgenic mice expressing human PrP 102L following the intracerebral inoculation of IPD P102L or classical CJD brain isolates ., The resultant prion isolates from these transgenic mouse brain were designated GSS-102L or CJD-102L prions , respectively , and because they are associated exclusively with disease-related conformers of human PrP 102L this enables unequivocal examination of the effects that this point mutation has on prion transmission barriers ., Our findings show that GSS-102L and CJD-102L prions are distinct from one another with divergent prion strain transmission properties following further passage in transgenic mice expressing either human PrP 102L or wild type human PrP ( Fig 4 ) ., Thus human PrP 102L is capable of supporting the propagation of distinct lethal prion strains and these data establish that the point mutation does not restrict PrP 102L to a single dominant pathogenic assembly state when templated by an exogenous prion strain ., Importantly our model has enabled us to isolate and investigate the transmission properties of prions originating from IPD P102L patient brain following amplification exclusively on human 102L PrP ., Our data show that 102L PrP prions from IPD P102L patient brain that generate PK-resistant PrP fragments of ~21–30 kDa have prion strain transmission properties distinct from all other prion strains propagated in acquired or sporadic human prion disease ., The most outstanding feature of this prion strain is its inability to propagate in transgenic mice expressing wild type PrP ., Significantly , the inability of GSS-102L prions to also propagate in wild type mice clearly shows that this prion strain is distinct from prions generated in IPD P102L prion-challenged 101LL PrP gene knock-in mice 36 ., The remarkable ease of transmission of 101L-passaged IPD P102L prions to wild type mice 36 contrasts strikingly with our data and suggests that a novel prion strain was propagated by the mutant mouse PrP rather than faithful replication of the authentic human PrP 102L prion strain ., We therefore recommend that future transgenic modeling of inherited prion disease should focus exclusively on using models that express the homotypic mutant human PrP primary sequence ., We and others have reported that variable involvement of disease-related conformers of wild type human PrP may contribute to phenotypic heterogeneity in IPD P102L 24 , 27 ., However the mechanism by which abnormal wild type PrP is deposited in P102L patient brain remains ill-defined ., Wild type PrP may be recruited by a seeded reaction with 102L PrPSc or may accumulate independently as a consequence of pathological changes associated with disease progression ., Notably , the glycoform ratios of proteinase K-resistant fragments of 102L PrP and wild type PrP from P102L patient brain are distinct from each other 24 27 suggesting that the 102L point mutation powerfully dictates thermodynamic preferences for disease-related PrP assembly states that cannot be adopted by wild type PrP and that a significant transmission barrier may be associated with conversion of wild type PrP by a 102L PrPSc seed ., This idea is supported by the observation that PK-resistant wild-type PrP in P102L patient brain does not appear to exceed approximately 10% of total PK-resistant PrP 24 , 27 ., Here we show that GSS-102L prions that propagate efficiently in further 102LL Tg27 transgenic mice fail to produce prion infection in transgenic mice expressing wild type human PrP ., Based upon the strength of this transmission barrier we conclude that seeded conversion of wild type PrP by abnormal conformers of 102L PrP that generate proteolytic fragments of ~ 21–30 kDa may , at best , be highly inefficient ., From these data it is tempting to speculate that abnormal conformers of 102L PrP that generate protease-resistant fragments of 8 kDa might instead be responsible for variable recruitment of wild type PrP in IPD P102L patient brain ., However other explanations may be equally possible ., In particular , our transmission experiments do not mirror the situation in IPD P102L patient brain where both PrP 102L and wild type PrP are co-expressed ., Thus in IPD P102L patient brain , wild type PrP will be exposed throughout the disease time course to all propagating 102L PrPSc species ( rather than only at inoculation ) and such prolonged exposure in vivo may be required for the generation of misfolded isoforms of wild type PrP ., Alternatively because prion strains appear to comprise a quasispecies maintained under host selection pressure ( rather than constituting a single molecular clone ) 2 , 48–50 minor subtypes of 102L PrPSc may be populated differently in individual P102L patients leading to variable degrees of recruitment of wild type PrP ., Notwithstanding such possibilities , at present we cannot conclusively resolve whether wild type PrP in IPD P102L patient brain misfolds through a directly seeded conversion reaction with an abnormal 102L PrP template or as a consequence of other pathological changes in the brain ., In this regard , transmission experiments in heterozygous transgenic mice expressing both 102L PrP and wild type PrP would not be able to differentiate between these possibilities ., Although the mechanism that leads to the accumulation of abnormal wild type PrP continues to remain ill-defined , this remains a potentially important contributor to phenotypic variation , not only in IPD P102L , but also in IPD associated with other PRNP mutations 51–56 ., Storage and biochemical analyses of post-mortem human brain samples and transmission studies to mice were performed with written informed consent from patients with capacity to give consent ., Where patients were unable to give informed consent , assent was obtained from their relatives in accordance with UK legislation and Codes of Practice ., Samples were stored and used in accordance with the Human Tissue Authority Codes of Practice and in line with the requirements of the Human Tissue Authority licence held by UCL Institute of Neurology ., This study was performed with approval from the National Hospital for Neurology and Neurosurgery and the UCL Institute of Neurology Joint Research Ethics Committee ( now National Research Ethics Service Committee , London—Queen Square ) —REC references: 03/N036 , 03/N038 and 03/N133 ., Work with mice was performed under approval and licence granted by the UK Home Office ( Animals ( Scientific Procedures ) Act 1986; Project Licence number 70/6454 ) and conformed to University College London institutional and ARRIVE guidelines ( www . nc3rs . org . uk/ARRIVE/ ) ., Transgenic mice homozygous for a human PrP102L , 129M transgene array and murine PrP null alleles ( Prnpo/o ) designated Tg ( HuPrP102L 129M+/+ Prnpo/o ) -27 mice ( 102LL Tg27 ) 33 have been described previously and were used without modification ., Transgenic mice homozygous for a wild type human PrP129M transgene array and murine PrP null alleles ( Prnpo/o ) designated Tg ( HuPrP129M+/+ Prnpo/o ) -35 congenic ( 129MM Tg35c ) were derived by subjecting previously described 129MM Tg35 mice 35 , 44 , 57 to commercially available speed congenic backcrossing on FVB/N genetic background ( Charles River UK ) ., Similarly , transgenic mice homozygous for a wild type human PrP129V transgene array and murine PrP null alleles ( Prnpo/o ) designated Tg ( HuPrP129V+/+ Prnpo/o ) -152 congenic ( 129VV Tg152c ) were derived by subjecting previously described 129VV Tg152 mice 37 , 39 , 42 to the speed congenic scheme ( Charles River UK ) ., Inbred FVB/NHsd mice were supplied by Harlan UK Ltd ., Strict bio-safety protocols were followed ., Inocula were prepared , using disposable equipment for each inoculum , in a microbiological containment level 3 laboratory and inoculations performed within a class 1 microbiological safety cabinet ., Ten mice per group from three transgenic lines , 102LL Tg27 , 129MM Tg35c , 129VV Tg152c and FVB/N wild type mice were inoculated with a panel of prion isolates , all previously passaged in 102LL Tg27 transgenic mice and therefore adapted to human 102L PrP ., The primary inocula comprised human brain homogenates from three IPD P102L patients , one sporadic CJD patient and three iatrogenic CJD patients ., Diagnosis of all cases had been neuropathologically confirmed ., The genotype of each mouse was confirmed by PCR of DNA prior to inclusion and all mice were uniquely identified by sub-cutaneous transponders ., Disposable cages were used and all cage lids and water bottles were also uniquely identified by transponder and remained with each cage of mice throughout the incubation period ., Care of the mice was according to institutional and ARRIVE guidelines ., Mice were anaesthetised with a mixture of halothane and O2 , and intra-cerebrally inoculated into the right parietal lobe with 30 μl of 1% ( w/v ) brain homogenate prepared in Dulbecco’s phosphate buffered saline lacking Ca2+ or Mg2+ ions ( D-PBS ) ., All mice were thereafter examined daily for clinical signs of prion disease ., Mice were killed if they exhibited any signs of distress or once a diagnosis of prion disease was established ., At post-mortem brains from inoculated mice were removed , divided sagittally with half frozen and half fixed in 10% buffered formol saline ., Anti-PrP monoclonal antibodies ICSM 18 and ICSM 35 were supplied by D-Gen Ltd , London , UK ., ICSM antibodies were raised in Prnpo/o mice against α or β PrP as described elsewhere 58 ., ICSM 18 is an IgG1 with an epitope spanning residues 142–153 of human PrP 58 ., ICSM 35 is an IgG2b with an epitope spanning residues 93–105 of human PrP 58 , 59 ., ICSM 18 recognizes both human PrP 102L and wild type human PrP whereas ICSM 35 recognizes wild type human PrP but not human PrP 102L 24 ., Brain homogenates ( 10% ( w/v ) ) were prepared in D-PBS and aliquots analysed in duplicate with or without proteinase K digestion ( 50 μg/ml final protease concentration , 1h , 37°C ) by electrophoresis and immunoblotting as described previously 60 , 61 ., Duplicate blots were blocked in PBS containing 0 . 05% v/v Tween-20 ( PBST ) and 5% w/v non-fat milk powder and probed with ICSM 18 or ICSM 35 anti-PrP monoclonal antibodies ( 0 . 2 μ g/ml final concentration in PBST ) in conjunction with anti-mouse IgG-alkaline phosphatase conjugated secondary antibody and chemiluminescent substrate CDP-Star ( Tropix Inc , Bedford , MA , USA ) and visualized on Biomax MR film ( Kodak ) as described 60 , 61 ., For analysis of PrP glycoforms , blots were developed in chemifluorescent substrate ( AttoPhos; Promega ) and visualized on a Storm 840 phosphoimager ( Amersham ) using ImageQuaNT software ( Amersham ) 31 , 61 ., Fixed brain was immersed in 98% formic acid for 1 h and paraffin wax embedded ., Serial sections of 4 μm nominal thickness were pre-treated with Tris-Citrate EDTA buffer for antigen retrieval 61 ., PrP deposition was visualized using ICSM 35 or ICSM 18 as the primary antibody , using an automated immunostaining system ( www . ventana . com ) ., Visualization was accomplished with diaminobenzidine staining ., Bright field photographs were taken on an ImageView digital camera ( www . soft-imaging . de ) and composed with Adobe Photoshop .

Introduction, Results, Discussion, Methods

Inherited prion disease ( IPD ) is caused by autosomal-dominant pathogenic mutations in the human prion protein ( PrP ) gene ( PRNP ) ., A proline to leucine substitution at PrP residue 102 ( P102L ) is classically associated with Gerstmann-Sträussler-Scheinker ( GSS ) disease but shows marked clinical and neuropathological variability within kindreds that may be caused by variable propagation of distinct prion strains generated from either PrP 102L or wild type PrP ., To-date the transmission properties of prions propagated in P102L patients remain ill-defined ., Multiple mouse models of GSS have focused on mutating the corresponding residue of murine PrP ( P101L ) , however murine PrP 101L , a novel PrP primary structure , may not have the repertoire of pathogenic prion conformations necessary to accurately model the human disease ., Here we describe the transmission properties of prions generated in human PrP 102L expressing transgenic mice that were generated after primary challenge with ex vivo human GSS P102L or classical CJD prions ., We show that distinct strains of prions were generated in these mice dependent upon source of the inoculum ( either GSS P102L or CJD brain ) and have designated these GSS-102L and CJD-102L prions , respectively ., GSS-102L prions have transmission properties distinct from all prion strains seen in sporadic and acquired human prion disease ., Significantly , GSS-102L prions appear incapable of transmitting disease to conventional mice expressing wild type mouse PrP , which contrasts strikingly with the reported transmission properties of prions generated in GSS P102L-challenged mice expressing mouse PrP 101L ., We conclude that future transgenic modeling of IPDs should focus exclusively on expression of mutant human PrP , as other approaches may generate novel experimental prion strains that are unrelated to human disease .

Inherited prion disease ( IPD ) is caused by pathogenic mutations in the human prion protein ( PrP ) gene leading to the formation of lethal prions in the brain ., To-date the properties of prions causing IPD and their similarities to prions causing other forms of human prion disease remain ill-defined ., In the present study we have investigated the properties of prions seen in patients with Gerstmann-Sträussler-Scheinker ( GSS ) disease associated with the substitution of leucine for proline at amino acid position 102 ( GSS P102L ) ., We examined the ability of these prions to infect transgenic mice expressing human mutant 102L PrP , human wild-type PrP or wild-type mice ., We found that GSS-102L prions have properties distinct from other types of human prions by showing that they can only infect transgenic mice expressing human PrP carrying the same mutation ., Mice expressing wild-type human PrP or wild-type mouse PrP were entirely resistant to infection with GSS-102L prions ., We conclude that accurate modeling of inherited prion disease requires the expression of authentic mutant human PrP in transgenic models , as other approaches may generate results that do not mirror the human disease .

journal.pcbi.1000008

Diminished Self-Chaperoning Activity of the ΔF508 Mutant of CFTR Results in Protein Misfolding

CF is the most common autosomal inherited disease with high morbidity among Caucasians ., CF patients have altered epithelial ion transport that leads to decreased hydration of epithelial surfaces in the gut , kidney , pancreas , and airways 1 ., Decreased surface liquid volume impairs mucociliary clearance which in turn leads to respiratory bacterial infection 2 , 3 ., Chronic pulmonary damage caused by bacterial infection dramatically decreases patients life expectancies ., The absence of a functional ABC protein , CFTR , from apical membranes of epithelial cells is the basis of this pathophysiology in cystic fibrosis 4 , 5 ., CFTR is a multidomain , integral membrane protein containing two transmembrane domains , two nucleotide-binding domains ( NBD1 and NBD2 ) , and a regulatory region ( R domain ) ( Figure 1 ) ., Although more than 1 , 400 mutations are known in CFTR ( http://www . genet . sickkids . on . ca/cftr ) , approximately 90% of CF patients carry the allele with deletion of the codon for phenylalanine at position 508 6 , which is located in the first nucleotide-binding domain ( NBD1 ) ( Figure 1 ) ., Experimental studies suggest that the CFTRΔF508 may be arrested at two stages during its biogenesis ., First , the loss of the Phe508 backbone may shift a fraction of the NBD1s of nascent CFTRΔF508 off the wild type folding pathway , causing misfolding and eventual rapid degradation 7–9 ., Interestingly , recent studies show no significant structural difference between the wild type and mutant NBD1 structures nor in their thermodynamic stabilities 10 ., Second , the absence of the Phe508 side-chain prevents the correct post-translational assembly of all CFTR domains 11 ., The detailed structural origin of the perturbed kinetics of NBD1 leading to its co-translational arrest is unknown ., Nucleotide-binding domains of ABC proteins are highly conserved in sequence and structure ., NBDs contain a typical F1 ATPase core subdomain , which consists of an α-helix surrounded by antiparallel β-sheets 9 , 12 ., This region contains the conserved Walker A and B motifs that are involved in binding ATP ., The α-helical subdomain contains the ABC-signature motif important for ATP hydrolysis ( Figure 1 ) ., From X-ray structures of bacterial transporters , the α-helical subdomain is also known to mediate contact with the transmembrane domains 13 , 14 ., Folding of multidomain proteins is aided by molecular chaperones to prevent and correct improper ( non-native ) associations between solvent-exposed hydrophobic regions ., Smaller single-domain proteins correct and prevent formation of improper contacts through a sequence of partial folding-unfolding events en route to the native state ., This sequence of partial folding-unfolding events reflects the ability of single-domain proteins to self-chaperone their folding ., In NBD1 , the attenuated refolding of the recombinant ΔF508 mutant is consistent with the notion that Phe508 reduces the activation energy of NBD1 folding in vivo as well as in vitro 11 ., Lowering of the activation energy increases the folding rate , which in turn reduces the folding time for NBD1 ., Reduction of the folding lessens the propensity of NBD1 to correct the malformed contacts in the intermediate states ., Here we propose that Phe508 deletion decreases NBD1s self-chaperoning capability ., To investigate the effect of the Phe508 deletion on the stability , dynamics and kinetics of NBD1 , we performed equilibrium dynamics simulations and folding simulations of NBD1-WT and NBD1-ΔF508 ., Our analysis shows that there is no significant difference in their stability and equilibrium dynamics , which agrees with experiments ., However , even in the presence of correcting mutants ( G550E , R553Q , and R555K ) 10 , 15–17 in our model of NBD1-ΔF508 , we still observe a significant change in dynamics at the folding transition ., We further explore the difference in the folding transition by performing 300 folding simulations each for NBD1-WT and NBD1-ΔF508 ., We also perform simulations of another mutant NBD1-F508A to serve as control ., These simulations allow the comparison of the mutant and wild type folding probabilities , their intermediate states , the structures of these intermediate states , and their folding pathways ., Finally , we identify contacts between residues in NBD1 critical to its folding dynamics that are perturbed by Phe508 deletion , thus increasing the propensity of NBD1-ΔF508 to misfold ., To determine the equilibrium dynamics and stabilities of the wild type and mutant NBD1 , we perform equilibrium simulations ( 106 time units∼0 . 5 millisecond 18 ) of wild type and mutant NBD1 using discrete molecular dynamics 19 , 20 ( see Methods ) ., From the equilibrium simulations , we calculate the thermal denaturation curve of both NBD1-WT and NBD1-ΔF508 ( Figure 2 ) and observe two stable thermodynamic states , folded and unfolded ., In agreement with previous experimental studies by denaturation experiments 7–9 , the stabilities of wild type and ΔF508 NBD1 are not significantly different ., The slope at the transition temperature of the wild type ( Tm∼0 . 68 ε/kB ) is 9838 kb and the slope at the transition temperature of the mutant ( Tm∼0 . 70 ε/kB ) is 16201 kb ( ε∼1–2 kcal/mol and kB is the Boltzman factor; see Methods for further discussion on units ) ., This shift in slope at the transition temperature indicates a difference in folding cooperativity of NBD1-WT and NBD1-ΔF508 and therefore a difference in folding kinetics ., Folding is a stochastic process , thus to investigate in detail the difference in folding kinetics and dynamics of NBD1-WT and NBD1-ΔF508 , we perform 300 folding simulations on each of the structures ., Starting from fully unfolded chains of NBD1-WTand NBD1-ΔF508 , we progressively reduce the temperature of the system to simulate thermal folding ( see Methods ) ., We find the folding probability 21 ( number of runs that lead to the native structure/number of total folding simulations ) of wild type to be 33±3% while that of the mutant is 13±2% ( see Methods ) ., The ratio of NBD1-WT and NBD1-ΔF508 correlates with the ratio of their folding yields derived from folding experiments ., Folding yields of NBD1-WT is approximately twice that of NBD1-ΔF508 in the temperature range 10°C to 22°C 9 ., Folding simulations of our control structure NBD1-F508A yield a folding probability of 26±4% which is intermediate to that NBD1-WTand NBD1-ΔF508 ., This folding probability value is in agreement with experimental studies showing intermediate folding efficiencies and maturation levels of NBD1-F508A relative to NBD1-WT 9 , 11 ., To investigate the molecular origin of the difference in folding yields and probabilities , we map the folding pathways of NBD1-WT , NBD1-F508A , and NBD1-ΔF508 by identifying their metastable folding intermediate states ., The folding intermediate states of a folding trajectory are exhibited as peaks in the energy probability distributions ( Figure 3; Figure S1 ) ., Thus , dominant intermediate states in the folding pathways are peaks in the average energy probability distributions ( see Methods; Figure 3 ) ., The average energy probability distributions of wild type and the mutant are significantly different ( Kolmogorov-Smirnov test; P-value<1 . 4×10−292 ) , which suggests a significant difference in the folding kinetics of wild type and mutant NBD1 ., The dominant intermediate states are listed in Table S1 ., The average fraction of native contacts of NBD1 structures in an intermediate state follows a distinct distribution ( Figure S2 ) , thus , an intermediate state identified using energy as the folding reaction coordinate , forms a distinct collection of NBD1 conformations ., We find that some intermediate states are accessible only by either NBD1-WT ( S6 and S9 ) or NBD1-ΔF508 ( S5 and S10 ) , further suggesting that Phe508 deletion leads the mutant to off-folding pathways ( see below ) ., While states S2 , S3 , S4 , S7 , and S8 are both traversed by NBD1-ΔF508 and NBD1-WT , their time occupancies ( length of time NBD1 spends in an intermediate state ) are different ( Figure 3B ) ., Since time occupancies are proportional to the free energy barriers between intermediate states , these observations suggest that the Phe508 deletion significantly perturbs the NBD1 folding free energy landscape ., To determine the difference between the sequence of folding events of the wild type , ΔF508 , and the F508A control , we estimate the probability of transitions between intermediate states ( see Methods and Figure S3 ) ., The difference in transition probabilities of NBD1-WT , NBD1-ΔF508 , and NBD1-F508A is shown in Figure 4 ., The transition probabilities show some states accessible only to either wild type or mutant NBD1 ., The difference in state accessibilities between the two indicates a difference in contact pattern formation ( nucleation events ) , which could cause the observed difference in folding yields ., We calculate the most dominant folding pathways in wild type and mutant NBD1 ., The most dominant path in wild type follows a sequence of transition Unfolded→S10→S8→S7→S5→S4→S1 , while the dominant path in the mutant follows the sequence of transitions Unfolded→S9→S8→S7→S6→S4→S1 ., Thus , NBD1-WTand NBD1-ΔF508 undergo different sequences of folding events ., Because of the reduction in dimensionality of the folding process when energy is used as a reaction coordinate , each intermediate state represents an ensemble of NBD1 structures ., To identify the primary structural characteristics of each intermediate state , we clustered structures in the corresponding state and calculated the frequency of contacts formed between pairs of residues ( Figure 5; see Methods ) ., In all intermediate states , we find the most notable structural difference between NBD1-WT and NBD1-ΔF508 occurs in the S7-H6 loop ., For example , P574 interacts with Q493 in wild type but not in the mutant ., Also , F575 interacts with F587 in the mutant but not in wild type ( Figure 6 ) ., This pattern of contact formation reflects the difference in NBD1-WT and NBD1-ΔF508 crystal structures that is embedded in the interactions defined according to structure ., Additionally , residue pairs that have similar interactions ( i . e . , attractive or repulsive ) in the wild type and mutant crystal structures still exhibit different contacts in the folding intermediate states ., These results show that the pattern of transient contact formation in the wild type is also perturbed by Phe508 deletion ., This class of residue pairs include Q525/E585 and C524/I586 ., We observe a number of folding trajectories reaching native energies ( ∼630 ε ) and within a 2 . 5 Å root-mean-square deviation ( RMSD ) with respect to the native structure , but the resulting topological wiring of the secondary structures is incorrect ., The “miswiring” consistently occurs in the H5-S6 loop ., Interestingly , this H5-S6 loop is in the immediate neighbourhood of the loop containing Phe508 ., This suggests “weak” regions in NBD1 that are intrinsically prone to misfolding ., To verify that the identified contact pairs ( Q493/P574 and F575/F587 ) found in the S7-H6 loop are indeed critical in the kinetics of NBD1 , we revert their interactions in NBD1-ΔF508 to their interactions in NBD1-WT and perform folding simulations ., In the case of the Q493/P574 pair , the residues are in close proximity in NBD1-WT but not in NBD1ΔF508 , thus we changed the interaction between Q493 and P574 in NBD1ΔF508 from repulsive to attractive to mimic a possible rescuing mutation ., Folding simulations of “rescued” NBD1-ΔF508 yield a folding probability of 19±2% ., On the other hand , residues F575 and F508 are in close contact in NBD1-ΔF508 but not in NBD1-WT , thus we reverted their interaction in NBD1-ΔF508 from attractive to repulsive ., Folding simulations of the second “rescued” NBD1-ΔF508 yield a folding probability of 20±2% ., These folding probabilities of the two “rescued” NBD1-ΔF508s are higher than the 13±2% folding probability of the original NBD1-ΔF508 , which supports our findings that the contacts between Q493 and P574 and between F575 and F587 are indeed critical to NBD1 folding ., Our results reveal the intrinsic property of NBD1ΔF508 to fold improperly and raise the possibility of redesigning NBD1ΔF508 to rescue it from misfolding ., In case of the contact that is found in wild type but not in the ΔF508 mutant ( e . g . , Q493/P574 ) , one can find amino acid substitutions that promote interaction between this pair of residues ( Q493/P574 ) ., On the other hand , for the contact found only in the ΔF508 mutant but not in wild type ( e . g . , F475/F587 ) , candidate rescue mutants are those that destabilize the interaction between this residue pair ( F475/F587 ) ., Knowing the molecular details of the altered folding in the case of the mutant domain also provides a basis for design of small molecules to correct the most prevalent and pathogenic mutation in CFTR ., To access time scales of NBD1 folding , we use a simplified protein model but still maintain important features of the protein such as side-chain packing ., Amino acid residues were modelled as follows: ( 1 ) glycines are represented by three beads ( -N , Cα , C′ ) ; ( 2 ) phenylalanine , tyrosine , tryptophan , and histidine by five beads ( -N , Cα , C′ , Cβ , Cγ ) , and ( 3 ) all other residues by four beads ( -N , Cα , C′ , Cβ ) 24 ., This protein model successfully described protein aggregation 24 ., In the simulations , we use PDB ID: 2BBO , PDB ID: 1XMI and PDB ID: 1XMJ 10 as models of NBD1-WT , NBD1-F08A , and NBD1-ΔF508 , respectively ., The missing loop between E403 and L436 in both wild type and mutant NBD1 is reconstructed using a loop-search algorithm in SYBYL ( Tripos Assoc . Inc , St . Louis , MO ) ., Using discrete molecular dynamics 19 , 20 , long equilibrium simulations at various temperatures were performed to investigate the equilibrium dynamics of the CFTR NBD1 ., Interactions between beads were defined using the Go̅-model 25 ., In the Go̅-model , interactions between residues are determined from the native structure of known NBD1 crystal structures ., Pairwise , square-well interactions were assigned between beads in the model according to contacts formed in the native state ., Specifically , two residues are said to be in contact if their atoms ( excluding hydrogen ) are within a distance of 4 . 5 Å ., The strength of the interaction between residues in contact ( denoted hereon as ε ) defines the energy units ., Physically ε∼1–2 kcal/mol , which is approximately a contribution to protein stability from a hydrogen bond ., The temperature is measured in units of ε/kb , where kb is the Boltzmann constant ., The time unit ( tu ) is estimated to be the shortest time between particle collisions in the system ( ∼0 . 1 nanosecond ) ., From long equilibrium simulations of 106 tu , we were able to access the long time-scale dynamics of the CFTR NBD1 in the order of 0 . 5 millisecond ., Each equilibrium simulation consumed approximately 300 CPU hours ., We perform 300 folding simulations for each NBD1-WT , NBD1-F508A , and NBD1-ΔF508 ., Starting from fully unfolded chains , the temperature of the system is progressively reduced to allow NBD1 to fold to its native structure ., Folding simulations proceeded until τmax∼60 , 000 tu ( time units ) , which is chosen to be longer than the typical folding time of the studied sequences 21 ., A similar criterion was employed in the studies calculating the folding probability of proteins 26 ., The NBD1 structure in a folding run is considered folded when ( 1 ) its energy is less than or equal to −620ε ( the energy of the native state ) , ( 2 ) its structure is within 2 . 5 Å RMSD from the native , and ( 3 ) the structure possesses correct topological wiring of the secondary structure elements ., To estimate the error in folding probabilities , each folding trajectory is considered a Bernoulli trial with a binary outcome , folded or unfolded ., The variance of a Bernoulli process is σ2\u200a=\u200ap ( 1−p ) /n , where p is probability and n is the total number of trials ., To identify the positions of intermediate states , a sum of multiple Gaussian curves is fitted to the average energy probability distribution of successfully folded runs ., ai , bi , and ci are the center , standard deviation and height of the ith Gaussian curve , respectively ., We estimate probability of transitions between states by counting the trajectories that underwent such transition ., The sum of probabilities of paths emanating from a given state is normalized to 1 , which physically means that the system always exits from its current intermediate state ., The transition probabilities represent independent conditional probabilities , thus the most likely path from the unfolded state to the native is estimated by multiplying the probabilities of the traced edges ., We calculated a contact matrix for each structure in the intermediate state ., An element of the contact matrix is 1 when two residues were within 4 . 5 Å or 0 otherwise ., Dominant contacts between pairs of residues in NBD1 are determined from the average contact matrix of all the structures .

Introduction, Results, Discussion, Methods

The absence of a functional ATP Binding Cassette ( ABC ) protein called the Cystic Fibrosis Transmembrane Conductance Regulator ( CFTR ) from apical membranes of epithelial cells is responsible for cystic fibrosis ( CF ) ., Over 90% of CF patients carry at least one mutant allele with deletion of phenylalanine at position 508 located in the N-terminal nucleotide binding domain ( NBD1 ) ., Biochemical and cell biological studies show that the ΔF508 mutant exhibits inefficient biosynthetic maturation and susceptibility to degradation probably due to misfolding of NBD1 and the resultant misassembly of other domains ., However , little is known about the direct effect of the Phe508 deletion on the NBD1 folding , which is essential for rational design strategies of cystic fibrosis treatment ., Here we show that the deletion of Phe508 alters the folding dynamics and kinetics of NBD1 , thus possibly affecting the assembly of the complete CFTR ., Using molecular dynamics simulations , we find that meta-stable intermediate states appearing on wild type and mutant folding pathways are populated differently and that their kinetic accessibilities are distinct ., The structural basis of the increased misfolding propensity of the ΔF508 NBD1 mutant is the perturbation of interactions in residue pairs Q493/P574 and F575/F578 found in loop S7-H6 ., As a proof-of-principle that the S7-H6 loop conformation can modulate the folding kinetics of NBD1 , we virtually design rescue mutations in the identified critical interactions to force the S7-H6 loop into the wild type conformation ., Two redesigned NBD1-ΔF508 variants exhibited significantly higher folding probabilities than the original NBD1-ΔF508 , thereby partially rescuing folding ability of the NBD1-ΔF508 mutant ., We propose that these observed defects in folding kinetics of mutant NBD1 may also be modulated by structures separate from the 508 site ., The identified structural determinants of increased misfolding propensity of NBD1-ΔF508 are essential information in correcting this pathogenic mutant .

Deletion of a single residue , phenylalanine at position 508 , in the first nucleotide binding domain ( NBD1 ) of the Cystic Fibrosis Transmembrane Conductance Regulator ( CFTR ) is present in approximately 90% of cystic fibrosis ( CF ) patients ., Experiments show that this mutant protein exhibits inefficient biosynthetic maturation and susceptibility to degradation probably due to misfolding of NBD1 and the resultant incorrect interactions of other domains ., However , little is known about the direct effect of the Phe508 deletion on NBD1 folding ., Here , using molecular dynamics simulations of NBD1-WT , NBD1-F508A , and NBD1-ΔF508 , we show that the deletion of Phe508 indeed alters the kinetics of NBD1 folding ., We also find that the intermediate states appearing on wild type and mutant folding pathways are populated differently and that their kinetic accessibilities are distinct ., Moreover , we identified critical interactions not necessarily localized near position 508 , such as Q493/P574 and F575/F587 , to be significant structural elements influencing the kinetic difference between wild type and mutant NBD1 ., We propose that these observed alterations in folding kinetics of mutant NBD1 result in misassembly of the whole multi-domain protein , thereby causing its premature degradation .

biophysics/protein folding

journal.pntd.0006566

Long-read whole genome sequencing and comparative analysis of six strains of the human pathogen Orientia tsutsugamushi

O . tsutsugamushi is an obligate intracellular bacterial pathogen of the order Rickettsiales , family Rickettsiaceae which causes the life-threatening human disease scrub typhus ., Orientia is transmitted by Leptotrombidium mites that occasionally feed on humans during the larval stage of development ( “chiggers” ) , inoculate bacteria into the skin , and initiate infection ., Orientia is maintained in mite populations by transovarial transmission ., The mites normally feed only once on a vertebrate host , and cannot transmit bacteria directly from one host to another 1 ., Bacteria propagate within endothelial cells , dendritic cells and monocytes at the site of inoculation , sometimes resulting in a visible red skin feature called an eschar 2 ., Bacteria subsequently spread through the endothelial and lymphatic systems to cause a systemic infection characterised by lymphadenopathy , headache , fever , rash and myalgia , which typically begin 7–10 days after inoculation ., The non-specificity of these symptoms makes scrub typhus difficult to diagnose based purely on clinical observations , and this is an important reason why the prevalence of scrub typhus has been historically under-recognised ., Scrub typhus has now been shown to be a leading cause of severe fever and sepsis in studies in Thailand , India , China , Laos and Myanmar 3 and untreated or severe cases are associated with CNS infection , morbidity and death 4 , 5 ., Locally acquired cases of scrub typhus have been reported in South America and the Middle East 6 , 7 , suggesting that the global burden of this disease may stretch beyond the traditionally known endemic areas of Asia and Northern Australia 3 ., O . tsutsugamushi is distinct from other members of the Rickettsiaceae ., The genus Orientia currently includes two known species , O . tsutsugamushi and O . chuto , the latter represented to date by a single strain inoculated from a patient with a febrile illness contracted in Dubai 6 ., High antigenic diversity among strains of O . tsutsugamushi is reflected in the poor immunological protection that recovered patients exhibit towards strains different from their original infection and this has hampered efforts towards vaccine development ., Despite its importance as a pathogen , few genomic analyses of O . tsutsugamushi have been published ., The first whole genome sequence , Boryong , 8 reported a proliferation of type IV secretion systems in a repeat-dense genome of which 37 . 1% comprised identical repeats ., A comparison of Boryong and the second complete genome , Ikeda 9 , revealed similar repeats present in each genome , dominated by an integrative and conjugative element named the O . tsutsugamushi amplified genetic element ( OtAGE ) , and identified a core genome of 520 genes shared between the two O . tsutsugamushi strains and the 5 available sequences of other Rickettsia 9 ., Extensive genomic reshuffling was thought to have been mediated by amplification of repetitive sequences ., In comparison to other Rickettsiae , many of which have small and extremely stable genomes , O . tsutsugamushi has a large genome with an extraordinary proliferation of repeat sequences and conjugative elements ., Some of the conjugative elements present in multiple copies across the genome are homologues of a gene cluster found in a single copy in Rickettsia bellii ., Many of the genes in these elements are fragmented , suggesting they are non-functional 10 ., Other intracellular pathogens also contain repetitive elements , such as the mobile genetic elements in Wolbachia 11 and the tandem intergenic repeats in Ehrlichia ruminantum 12 ., These mechanisms may evolve to increase genetic variability and aid immune evasion in bacteria which cannot easily take up novel DNA ., Larger collections of O . tsutsugamushi strains have been extensively studied using MLST and sequence typing of the groES and groEL 13 genes , and the outer membrane proteins 47kDa ( also called HtrA or TSA47 ) 14 and 56kDa ( also called OmpA or TSA56 ) 15 genes ., The 56kDa and 47kDa genes are highly immunogenic in human patients and animal models and have long been investigated as candidates for vaccine design , but high levels of diversity between strains , especially in the 56kDa gene , have limited the potential of developing a universal vaccine based on these epitopes ., Multiple studies in South East Asia have looked at the diversity of strains by MLST and 56kDa typing , and shown a high level of diversity , with many MLSTs unique to an individual strain 16–19 ., Work in Thailand and Laos has shown recombination between MLSTs , as well as evidence for multiple infections in individual patients , implying that multiple strains may co-exist in mites 16 ., Comparisons of 56kDa typing with MLST 16 and 47kDa 14 also show low congruence between methods , suggesting that single gene typing of Orientia may not represent the true relationships between strains; by extension , a 7-gene MLST scheme may not capture the full set of genomic relationships among strains ., Attempts to generate complete O . tsutsugamushi genomes by whole genome sequencing have been limited by the difficulties of sequencing and assembling a repeat-dense genome , and no further genomes have been completed since the Boryong and Ikeda genomes in 2008 ., Current draft assemblies are fragmented with over 50 contigs per genome , and vary in size–the two assemblies of the genome of O . tsutsugamushi str ., Karp available on Genbank are 1 , 459 , 958bp ( https://www . ebi . ac . uk/ena/data/view/LANM01000000 ) and 2 , 022 , 909bp ( https://www . ncbi . nlm . nih . gov/nuccore/LYMA00000000 ) 20 in length , suggesting that assemblies are either incomplete , or have problems caused by the misassembly of repeats or the inclusion of contaminating sequences ., In this work , we have used Pacific Biosciences long-read sequencing to assemble six complete genomes of O . tsutsgamushi strains representing a range of geographical origins and serotypes ., This expanded genomic perspective will contribute to our understanding of the phylogeography and epidemiology of this species , as well as contribute to more detailed studies of virulence mechanisms ., All experiments were performed using O . tsutsugamushi grown in the mouse fibroblast cell line L929 ( European Collection of Authenticated Cell Cultures 85103115 ) ., Uninfected L929 cells were grown in 25 cm2 and 75 cm2 plastic flasks at 37 oC and 5% CO2 , using DMEM or RPMI 1640 ( Thermo Fisher Scientific , USA ) media supplemented with 10% FBS ( Sigma ) as described previously ( Giengkam et al 2015 ) ., Infected L929 cells were grown in the same way , but at 35 oC ., Frozen stocks of bacteria were grown for 5 days , then the bacterial content was calculated using qPCR against the bacterial gene TSA47 21 ., Bacteria were isolated onto fresh L929 cells in 75 cm2 flasks at an Multiplicity of Infection of 10:1 and then grown for an additional 7 days ., At this point bacteria were isolated from host cells and prepared for DNA extraction ., The supernatant were removed from infected flasks and replaced with 6–8 ml pre-warmed media ., Infected cells were harvested by mechanical scraping and then lysed using a bullet blender ( BBX24B , Bullet Blender Blue , Nextadvance , USA ) operated at power 8 for 1 min ., Host cell debris was removed by centrifugation at 300xg for 3 minutes , and the supernatant was filtered through a 2 . 0 μm filter unit ( Puradisc , GE Whatman , USA ) ., 10 μl of 1 . 4 μg/μl DNase ( Deoxyribonuclease I from bovine pancreas , Merck , UK ) was added per 1 ml of bacterial solution , then incubated at room temperature for 30 minutes ., This procedure removed excess host cell DNA ., The bacterial sample was then isolated by centrifugation at 14 , 000xg for 10 min at 4 oC , and washed two times with 0 . 3M sucrose ( Merck ) ., After the washing steps were completed DNA was extracted using a QIAGEN DNeasy Blood & Tissue Kit ( QIAGEN , UK ) following the manufacturer’s instructions ., Purified DNA samples were analysed by gel electrophoresis using 0 . 8% agarose gel , in order to assess the DNA integrity ., The yield of genomic DNA was quantified using a nanodrop ( Nanodrop 2000 , Thermo Scientific , UK ) and Qubit Fluorometric Quantitation ( Qubit 3 . 0 Fluorometer , Thermo Scientific ) ., SMRTBell templates were prepared from purified Orientia genomic DNA according to PacBio’s recommended protocols ( PacBio , USA ) ., Briefly , 20kb libraries were targeted; enrichment for large fragments was done using BluePippin ( Sage Science , USA ) size selection method or successive Ampure ( Beckman Coulter , USA ) clean-ups , depending on the original DNA size distribution and quantity , as recommended by PacBio ., SMRTBell templates were sequenced on a Pacific Biosciences RSII Sequencer using P6 chemistry with a 240min run time ., Genomes were assembled using the RS_HGAP_Assembly . 3 protocol from the PacBio SMRTPortal ( version 2 . 3 . 0 ) , with initial polishing performed on trimmed initial assemblies using the same raw sequencing data with the RS_Resequencing . 1 protocol ., Each assembly was further polished using paired-end reads sequenced on an Illumina Miseq machine ., Sequencing information and data availability for each sample can be found in S1 Supporting Information ., For each assembly , the corresponding Illumina reads were aligned to the PacBio assembly using Stampy v1 . 0 . 23 22 ., Pilon v1 . 16 23 was then used to generate a final genome , and corrected 2 to 265 errors in the assemblies , with the majority of the errors being single base deletions at the end of A or T homopolymer runs ., All genomes were rotated and reverse complemented as needed so that the predicted start codon for the dnaA gene formed the first nucleotide in the genome sequence ., Sequencing and assembly statistics can be found in Table 1 and S1 Supporting Information ., The Boryong , Ikeda , and non-Orientia Rickettsial genomes used in this study were obtained from NCBI ( S1 Supporting Information ) ., The finished assemblies were annotated using Prokka v1 . 11 24 , using a custom database created from the Boryong and Ikeda strains , which were previously annotated using the NCBI prokaryotic annotation pathway ., The Boryong and Ikeda strains were re-annotated using Prokka for consistency with the other samples ., Short gene names for all non-hypothetical gene products were checked manually ( 607 products ) ., Where genes names were present for Boryong and/or Ikeda a consensus name based on these was selected ., Where no short name was available , the long gene name was searched for in E . coli using the UniProt database , and where a single and unambiguous match was selected this was used ., In cases of ambiguity the protein sequence from Orientia was used in a BLAST search against E . coli , R . rickettsii and H . sapiens and the short name of the closest match was selected ., The key Orientia genes TSA56 , TSA47 , TSA22 , ScaA , ScaC , ScaD , and ScaE were also manually annotated by taking known protein sequences from the UT76 strain and using BLAST to find the homologous genes in the other strains and give them the correct names ., Repetitive regions of the genome were defined as regions of at least 1000bp in length which had a match with another 1000bp region with up to 100 differences ( mismatches , insertions , and deletions ) allowed ., The repetitive regions were identified with Vmatch 25 ., The core and accessory genome was identified using Roary 26 with a threshold of 80% sequence identity required to consider two sequences part of the same gene group ., Core genes were defined as genes present in every sample and as a single copy in every sample ., The accessory genes identified using Roary were re-clustered using CD-Hit 27 , 28 with a cutoff of 80% identity across 95% of the length of the shortest protein to identify accessory genes which were truncated copies of other proteins ., The correlation between gene order in each pair of samples was calculated by taking the order of the genes relative to the Karp strain and calculating the Spearman’s rank coefficient between each pair ., COG categories were assigned using RPS-BLAST to find matches in the NCBI Conserved Domain Database 29 and assigning a COG category to these using cdd2cog 30 ., Repeat genes were identified using protein clusters generated by CD-Hit to find gene groups which were present at more than 1 copy ., CD-Hit was used to identify protein clusters on the proteins predicted by Prokka with a cutoff of 80% identity across 90% of the length of the shortest protein ., Pseudogenes were identified from CD-Hit protein clusters where at least one protein was a truncated version of the longest protein in the group ., As pseudogenes which are truncated at the 5’ end will not be annotated by Prokka , tBLASTn 31 was using to screen for any additional pseudogenes in non-genic regions by searching for BLAST hits with protein identity > = 80% and an E-value <10−15 ., This method found a further 26–37 pseudogenes per strain ., Further analysis used BioPython 32 and the GenomeDiagram package 33 ., Fig 1 was created with Circos 34 ., Statistical tests were carried out in R 35 and the Python SciPy library 36 ., Phylogenies were inferred using Maximum Likelihood methods in RaxML 37 under the GAMMA model of rate heterogeneity and bootstrap values calculated using the rapid bootstrap method ., The input sequences were aligned with Clustal Omega 38 ( for the 56kDa/46kDa/MLST trees ) or using the MAFFT alignments produced by Roary ( for the core gene tree ) ., Phylogenetic trees were drawn using the ape 39 and phytools 40 R packages , and Robinson-Foulds distances were calculated using the phangorn 41 R package ., Eight genomes ( Table 1 ) were assembled using the PacBio reads to perform initial genome assembly and Illumina sequencing reads to polish the genomes and reduce errors ., Six of the eight genomes could be assembled into a single finished contig , while two genomes remain in multiple contigs ( Table 1 ) ., In addition , two previously assembled references genomes , O . tsutsugamushi str ., Boryong and O . tsutsugamushi str ., Ikeda , were incorporated into our analysis ., The genome size ranges from 1 . 93Mb to 2 . 47Mb , and the GC content for all strains is consistent at 30–31% ., We assessed the genomes to identify core genes shared between all genomes , and look for repetitive regions and repeat genes in each strain ., Fig 1 plots the genetic elements of each complete genome ., The number of predicted genes in each strain ranges from 2086 ( UT176 ) to 2709 ( Gilliam ) ( Table 2 ) and is highly correlated with genome size ( Spearman’s correlation coefficient 0 . 94 , p < 2 . 2x10-16 ) ., A function could not be assigned , by similarity to reported sequences , to 325–547 genes ( 16 to 22% of the identified coding regions ) in each strain ., The set of 8 complete , single-contig genomes was used to identify core genes ( present in all genomes ) and accessory genes ( present in a subset of genomes ) , using the criterion that all members of a group of putative orthologues should be at least 80% identical to all other members of the group ., The number of predicted genes and total genome size for the unfinished genomes was comparable with those in the finished genomes , which suggests that while the assemblies are not assembled into a single contig they are not missing large pieces of the genome sequence of these strains ., However , since we are unable to verify that they contain the complete genome , we have excluded them from the core gene analysis to avoid concluding that genes are not in the core genome based on their absence from these assemblies ., A total of 657 genes were present in all 8 strains and therefore form a putative core genome , while 2812 gene clusters ( which may contain multiple genes from a single strain ) were present in 2–7 of the 8 strains , and a further 4687 gene clusters were found in a single strain ., The 657 core genes make up 28–35% of the genome of each strain ( S1 Supporting Information ) ., The number of core genes does not continue to decrease as more genomes are added to the analysis , suggesting that the core genome of Orientia can be defined with 8 representative genomes ( Fig 2 ) ., In the initial analysis with Roary , the total number of gene clusters continued to grow , suggesting an open pan-genome ., However , while the total number of distinct gene clusters continues to grow , many of these clusters contain genes which are annotated with the same function as previously observed gene clusters ., Of the 6050 clusters where a function can be assigned to one or more gene , there are only 122 different predicted gene products , many of them conjugal transfer proteins , transposases , DNA helicases , and other functions shared by genes known to be part of the O . tsutsugamushi amplified genetic element identified in the Ikeda strain 42 ., Re-clustering these accessory genes using a more lenient length cut-off to determine clusters allows genes which are only a match to part of a gene sequence to cluster together to include more truncated and fragmented copies of genes ., A comparison between the standard Roary clusters and this lenient clustering shows that the number of accessory gene clusters continues to increase , but at a slower rate ( S1 Supporting Information ) ., The number of gene products remains constant at 122 no matter how many strains are included in the analysis ., This suggests that the increase in non-core gene clusters is mainly due to further duplication and truncation of existing genes , rather than by the import of novel genes ., With the completed genomes produced by long read sequencing , the synteny of the genomes can be investigated ., Previous work on the Boryong and Ikeda genomes showed extensive genome shuffling between the two strains ., Analysis of the order and grouping of the core genes which are conserved in each genome shows that the genome has undergone massive rearrangement , with the core genes found in ‘core gene islands’ with repeat regions interspersed between these islands ., The 657 core genes are present in 145–157 distinct islands , of which only 51 are conserved ( defined as the same genes present in the same order , including single member islands ) in all genomes ., Fig 3 shows the position and ordering of these conserved core gene islands which are maintained in all samples relative to the position and ordering in the Karp strain ., The correlation between gene order in each pair of samples is shown in S1 Supporting Information ., A value close to 0 shows low correlation in gene order , while values closer to 1 show higher correlation in gene order ., As there are differences in the correlation of gene order between strains , this suggests that the process of genome rearrangement is happening in multiple steps and not as a single event ., The identities of genes present on conserved islands is shown in S1 Supporting Information ., Conserved islands range from 1–13 genes in size , with larger islands often containing genes linked by plausible biological functions ., For example , groups 3 and 6 include a number of cell division and peptidoglycan biosynthesis genes ( including mraY , murF , murE , pbp , ftsL , dnaJ and dnaK in group 3 and murC , murB , ddl and ftsQ in group 6 ) and groups 31 and 32 include a number of 30S and 50S ribosomal proteins ., Analysis of the number of conserved islands shared between samples shows that the number of conserved islands continues to decrease as more genomes are included ( Figure S3 in S1 Supporting Information ) , and suggests that gene order and clustering is not always constrained in O . tsutsugamushi ., There is no difference seen in the size of the islands between conserved and non-conserved islands ( Figure S4 in S1 Supporting Information ) ( two-sample Kolmogorov-Smirnov test D = 0 . 085 , p-value = 0 . 86 ) , the nucleotide diversity between genes in the two categories of islands ( two-sample Kolmogorov-Smirnov test D = 0 . 052 , p-value = 0 . 86 ) ( Figure S5 in S1 Supporting Information ) , or the Clusters of Orthologous Groups ( COG ) categories assigned to genes in the two island categories ( Chi-squared test χ2 = 15 . 03 , p = 0 . 82 ) ( Figure S6 in S1 Supporting Information ) ., The genomes of O . tsutsugamushi are known to be highly repetitive , including a highly amplified genetic element known as the O . tsutsugamushi amplified genetic element ( OtAGE ) , as well as other transposable elements ., Our results emphasise the large number of repeated genes and regions , including many genes related to the Type IV secretion system ., The total proportion of the genome which is repetitive ( see Methods for our definition of repetitive ) differs markedly from 33% in UT176 to 51% in Gilliam ( S1 Supporting Information ) ., In contrast , the extremely compact ( and therefore non-repetitive ) Rickettsia typhi genome is 0% repetitive by our measure and even , intriguingly , the Rickettsia endosymbiont of Ixodes scapularis , known to encode multiple copies of the same repetitive element found in Orientia 43 , is 20% repetitive in our analysis , despite our methods giving more conservative figures than previously determined for the Ikeda strain 42 ., We identified 530 groups of repeat genes containing 12043 genes present in multiple copies in at least one strain ., Of the 530 groups , 427 represent genes found in multiple strains , while 103 are found only in a single strain ., Despite clustering in 530 groups , the genes have only 66 different functional products , as is expected from the earlier results looking at all the non-core genes ., The repeat genes are mainly transposase and conjugal transfer genes , similar to those previously reported in the OtAGE ( S1 Supporting Information ) , and cluster into genetic elements which are interspersed between the core genes ., Many of these genes are present in high copy number , with all strains carrying over 200 transposases and 300 conjugal transfer genes and gene fragments ., These 530 repeat elements occupy 35–47% of the O . tsutsugamushi genome and represent 57–67% of the genes in these genomes ( S1 Supporting Information ) ., O . tsutsugamushi genes are known to exhibit high levels of pseudogenisation and gene decay ., We searched for pseudogenes in each genome , and identified up to 484 pseudogenes per strain ( S1 Supporting Information ) ., This is lower than previously reported in Ikeda , but due to methodological differences the figures cannot be directly compared ., We also assessed whether the pseudogene had been caused by truncation at the 5’ or 3’ end of the sequencing , or by frameshift ., We did not see a larger number of pseudogenes caused by frameshifts in the genomes new to this study compared to the Boryong and Ikeda strains , suggesting that we do not have a large number of frameshift errors caused by PacBio sequencing ., A phylogenetic tree was constructed using the core genes from each strain ., This can be compared to trees built using the 56kDa ( Fig 4 ) and 47kDa ( Figure S7 in S1 Supporting Information ) genes , which are often used for phylogenetic analysis of O . tsutsugamushi , or to trees built using the 7 MLST genes from the MLST scheme developed in Sonthayanon et al . , 2010 ( Figure S8 in S1 Supporting Information ) ., Orientia strains are commonly based on their similarity to reference strains , either from phylogenetics or serology ., Compared to the 56kDa tree , the core gene tree suggests the Kato and Ikeda strains are more closely related to the Karp , UT176 , and UT76 strains than the TA686 and Gilliam strains ( Fig 4 ) ., Robinson-Foulds distances between trees are shown in S1 Supporting Information; for this small number of strains , the distance is lowest between the 47kDa tree and the core genome tree ., We present the first large-scale genomic comparison of O . tsutsugamushi , a bacterium which is important both for the study of human disease and for its unique insights into genome evolution ., Previous studies of O . tsutsugamushi genomes have used BAC cloning and Sanger sequencing to produce complete genomes 8 , 42 , or have used next-generation sequencing strategies which have produced only incomplete and fragmented genomes 20 ., We demonstrate that a combination of PacBio and Illumina sequencing is sufficient to produce a single-contig genome , allowing us to study the gene content and synteny of this organism ., For the two genomes which could not be assembled into single contigs in our study ( FPW1038 and TA763 ) , we found that the sequencing produced fewer reads at the high end of the length distribution ., This suggests that given the highly repetitive nature of the O . tsutsugamushi genome , the DNA preparation and sequencing methods must be carefully chosen to produce very long reads in order to produce complete assemblies ., We used Illumina sequencing to correct errors in our genomes , which was vital to reduce the number of homopolymer errors , which could otherwise suggest frameshift errors and affect gene annotation ., While the fewest errors we corrected in a strain was two , this is likely an underestimate as no Illumina reads map with high confidence to 5–15% of the genome due to the repeats , and regions where Illumina reads cannot map will be impossible to correct using this approach ., While our analysis shows small differences when quantifying the extent of the repeat regions and repeat gene families in Orientia compared to previous work , a direct comparison is difficult due to differences in methodology between analyses ., Owing to the difficulties of producing complete genomes , most previous work has relied on single gene or MLST studies to investigate the genetic diversity of O . tsutsugamushi ., We demonstrate that phylogenies generated from limited data are substantially different from those produced from the whole core genome ., The common practice of grouping Orientia strains into ‘Karp-like’ or ‘Gilliam-like’ groups based on the genotype of the 56kDa antigen may not give an accurate view of the relatedness of these strains , especially when recombination is taken into account , although this may still be important when considering immune response ., The core gene set we have defined can be sequenced using Illumina reads alone and will allow future studies to perform large-scale phylogenetic analysis of Orientia without needing long-read sequencing ., We have not investigated the effect of recombination in this study due to the difficulties of analyzing recombination in a highly fragmented and rearranged genome , and it is possible that the core gene tree we present does not represent the true phylogeny of these strains ., Previous work has demonstrated limited synteny between the two reference strains of O . tsutsugamushi , but we extend this to demonstrate that there is minimal synteny between any known O . tsutsugamushi genome ., The pattern of core gene islands separated by transposable elements and repeats suggests a repeat-mediated system of chromosome rearrangement ., It is unclear whether this is a gradual process of genome rearrangement , or whether the genome is being broken apart and rearranged rapidly , similar to chromothripsis or the chromosome repair of Deinoccocus radiodurans after exposure to ionizing radiation ., In Deinococcus , it is thought that RecFOR pathway is particularly important for DNA repair , and it has no homologues to RecB or RecC 44 ., Similarly , in Orientia , the core genome does not contain RecB or RecC , but does contain the RecFOR pathway genes , indicating this alternative DNA repair pathway may also be important ., Longitudinal studies of O . tsutsugamushi genomes during passage or infection may be needed to determine the speed and processes of genome rearrangement in Orientia ., We report a core genome of only 657 genes , compared to the 519 previously reported as the core genome shared between Orientia and five other sequenced Rickettsia , but similar to the 665 core genes shared between the Ikeda and Boryong strains in a previous study 9 ., Differences in methodology may lead to the reporting of different core gene sets , but more interesting is the pattern of core genome islands separated by amplified repeat regions , and the lack of conservation in the ordering and clustering of the core genes ., All of the Orientia genomes show high repetitiveness , which we measured as both non-unique regions of the genome , and genes which are present in multiple copies ( some of which may be truncated ) ., The genomes of intracellular bacteria tend towards genome reduction and gene loss 10 , 45 , but maintain degraded genes and accumulate non-coding DNA ., The transition to intracellularity leads to smaller effective population sizes , as the bacteria are sequestered inside cells 46 ., As the majority of mutations have a neutral of slightly deleterious effect , they will be removed by purifying selection; however , purifying selection is less effective in smaller bottlenecked populations such as intracellular bacteria , which will lead to an accumulation of slightly deleterious mutations and an increased rate of sequence evolution 47 ., The expansion of the OtAGE ( and other mobile elements ) throughout the Orientia lineage appears to be another consequence of the decreased role of selection on Orientia in its intracellular niche , again leading to accelerated sequence evolution of the genome through rearrangement and gene loss ., This is supported by the finding that the diversity of gene repertoire between strains of O . tsutsugamushi is partly due to the duplication and truncation of existing genes ., The amplification of a transposable element has been seen in Rickettsial 43 and non-Rickettsial 48 species , but it is not known whether this is associated with rearrangement of the genome in other species ., High levels of genome plasticity and recombination have also been seen in fungal endobacteria , and are thought to be a defence against Muller’s ratchet 49 In conclusion , we report the generation of six complete and a further two draft genomes from a diverse set of strains of the important but neglected human pathogen O . tsutsugamushi ., This set includes the major reference strains Karp , Kato and Gilliam , and will serve as a valuable resource for scientists and clinicians studying this pathogen , in particular supporting future work on Orientia genomics , vaccine development , and cell biology ., The new genomes reported here confirm the status of Orientia as one of the most fragmented and highly repeated bacterial genomes known , and exciting questions remain regarding the mechanisms and timeframes driving the evolution of these extraordinary genomes .

Introduction, Methods, Results, Discussion

Orientia tsutsugamushi is a clinically important but neglected obligate intracellular bacterial pathogen of the Rickettsiaceae family that causes the potentially life-threatening human disease scrub typhus ., In contrast to the genome reduction seen in many obligate intracellular bacteria , early genetic studies of Orientia have revealed one of the most repetitive bacterial genomes sequenced to date ., The dramatic expansion of mobile elements has hampered efforts to generate complete genome sequences using short read sequencing methodologies , and consequently there have been few studies of the comparative genomics of this neglected species ., We report new high-quality genomes of O . tsutsugamushi , generated using PacBio single molecule long read sequencing , for six strains: Karp , Kato , Gilliam , TA686 , UT76 and UT176 ., In comparative genomics analyses of these strains together with existing reference genomes from Ikeda and Boryong strains , we identify a relatively small core genome of 657 genes , grouped into core gene islands and separated by repeat regions , and use the core genes to infer the first whole-genome phylogeny of Orientia ., Complete assemblies of multiple Orientia genomes verify initial suggestions that these are remarkable organisms ., They have larger genomes compared with most other Rickettsiaceae , with widespread amplification of repeat elements and massive chromosomal rearrangements between strains ., At the gene level , Orientia has a relatively small set of universally conserved genes , similar to other obligate intracellular bacteria , and the relative expansion in genome size can be accounted for by gene duplication and repeat amplification ., Our study demonstrates the utility of long read sequencing to investigate complex bacterial genomes and characterise genomic variation .

Orientia tsutsugamushi is an obligate intracellular bacterial pathogen , and the causative agent of the human disease scrub typhus ., This vector-borne bacterial infection can be life-threatening if not treated rapidly with appropriate antibiotics ., Orientia is endemic across large parts of Asia and is estimated to infect at least one million people per year ., O . tsutsugamushi has a highly unusual genome ., At 2 . 1 Mbp in length , it is almost double the size of most other obligate intracellular bacterial genomes , which have undergone extensive genome reduction in their adaptation to the intracellular lifestyle ., The comparatively expanded genome of Orientia is accounted for by an abundance of repeated DNA elements ., Indeed , the genome of Orientia is the most highly repeated bacterial genome reported to date ., As a consequence of this , it has been difficult to generate complete genome assemblies of Orientia using short-read sequencing technologies ., In the current report , the authors use Pacific Biosciences long-read sequencing to generate six complete genomes of diverse strains of O . tsutsugamushi ., By comparing these genomes , the authors show that there is very little colinearity in the ordering of genes within the genome , even between closely related strains ., This lack of synteny suggests that there have been frequent events of massive chromosomal rearrangement in this species , although the underlying mechanisms for this have not been shown ., The genome sequences and analysis reported here will benefit those working on various aspects of the biology of this clinically important but neglected bacterial species .

bacteriology, sequencing techniques, geomorphology, taxonomy, medicine and health sciences, pathology and laboratory medicine, landforms, intracellular pathogens, pathogens, topography, microbiology, genome sequencing, phylogenetics, data management, phylogenetic analysis, genome analysis, bacterial genetics, molecular biology techniques, microbial genetics, bacterial pathogens, microbial genomics, islands, research and analysis methods, orienta tsutsugamushi, bacterial genomics, genomic libraries, computer and information sciences, medical microbiology, microbial pathogens, comparative genomics, evolutionary systematics, molecular biology, earth sciences, genetics, biology and life sciences, genomics, evolutionary biology, computational biology

journal.pcbi.1000446

Identification of a Kinase Profile that Predicts Chromosome Damage Induced by Small Molecule Kinase Inhibitors

Toxicity is a major cause of attrition in drug development ., While identifying liabilities and potential toxicity is difficult and costly , safety issues can become markedly more complex when kinases are the pharmaceutical target ., Kinases regulate many basic functions in normal cells ., When their activity is altered , kinases can be the mechanistic reason for a cell to acquire an abnormal phenotype ., In metabolic , oncologic , viral , cardiovascular and inflammatory diseases , over 150 different kinases , of the over 500 known protein kinase family members , are considered putative drug targets 1 ., Marketed small molecule kinase inhibitors ( SMKIs ) have suitably demonstrated the effectiveness of this therapeutic approach for oncologic indications 2 ., SMKIs intended for non-oncologic diseases , however , are increasingly represented in various stages of preclinical and clinical development 1 ., Most SMKIs exert their pharmacologic effect by interacting with the ATP binding pocket 3 , inhibiting the ability of the kinase to phosphorylate the intended substrate , and blocking downstream signal transduction ., Because of the evolutionarily conserved nature of the ATP binding pocket , a SMKI intended to inhibit a particular kinase may potently inhibit dozens of other kinase members across the human kinome 4 ., Off-target kinases can be a potential safety liability of this therapeutic class and hinder drug development ., The mechanisms by which different toxicities arise as a result of off-target inhibition are not well characterized ., Sutent , a highly non-selective inhibitor of multiple tyrosine kinases and Gleevec , a relatively selective Bcr-Abl inhibitor , both increase the risk of cardiotoxicty 5–7 , though additional , less publicized toxicities , are also common for SMKIs ., Kinases are key regulators of mitosis , as they are intricately involved with precise signaling and the coordination needed for proper replication and segregation of chromosomes into daughter cells 8–10 ., While kinases may be targeted for their role in pathways associated with a disease of interest , inhibition of kinases may also disrupt normal cellular processes ., A frequently observed toxicity for SMKIs is a positive result for chromosomal damage in an assay of DNA integrity , which likely occurs as the result of inhibiting kinases involved in mitosis or chromosomal segregation ., The micronucleus test ( MNT ) is widely regarded as a sensitive assay for genetic toxicity as it is a means to detect either pieces and/or whole chromosomes that appear as a micronucleus in the daughter cell following chemical exposure 11 , 12 ., A positive result in this assay can hamper or halt drug development , as it is a biomarker of chromosomal damage , which is a hallmark of cancer 13–15 ., Thus , human exposure to aneugens or clastogens should be attenuated , or avoided altogether , when possible ., A small number of kinases , such as the polo-like and aurora kinases 16–19 , are known to associate with chromosomal damage , however the genotoxic potential associated with inhibiting the majority of the kinome is largely unknown ., MNT results can be considered a surrogate , and sometimes predictive endpoint for carcinogenicity ., This study models this endpoint because of its correlation with genotoxicity and the availability of a set of training compounds that have been screened with this assay ., Although regulatory agencies require such an in vitro assay prior to moving forward with preclinical development , there are advantages to modeling this assay in silico ., Namely , because of the low-throughput nature of the assay , the drug discovery process would benefit from a cheaper , faster screen that could assist in reducing the number of leads that typically fail at later stages , as well as help design compounds with fewer safety liabilities ., Since all promising SMKIs at Roche are tested in kinase inhibition assays , these data present the opportunity to explore possible correlations between SMKI kinase selectivity and the potential to cause chromosomal damage ., The objective of this study was to identify kinases that correlate with chromosomal damage when inhibited ., At Roche , we aimed to use these findings as a set of kinases that medicinal chemists should avoid when designing compounds , so as to avoid positive MNT results , thereby reducing attrition rates ., By using machine learning methods on data that were already available from early kinase-based high-throughput screens , we were able to identify such a set of kinases and develop a fast and efficient model for predicting whether a compound will test positive for genotoxicity ., Besides its novel utility in the drug discovery pipeline , the model also sheds light on the biological mechanisms of genotoxicity , and allows us to create hypotheses for further studies ., The 113 SMKIs were chosen to represent a diversity of compound properties and structural moieties ., Figure 1 shows a Principal Components Analysis ( PCA ) plot of the training compounds , in color , overlaid on top of a plot of all Roche compounds that have been screened with Ambit Biosciences ( San Diego , CA ) KINOMEscan assay of 317 kinases ., The PCA was based on structural fingerprints , a representation of the molecular structure of each compound ., This method of analysis is a means for reducing dimensionality to best explain variability in the data ., Figure 1 shows that structures of the 113 compounds are highly variable and sample the chemical space of the entire Roche SMKI library well ., With a diverse training set , chances of redundancy are reduced and the model is likely to be more robust to future predictions ., During dataset preprocessing , 27 mutant kinases were removed from the initial panel of 317 kinases as their inclusion and potential selection would be difficult to interpret from biological and mechanistic standpoints ., This yielded a panel of 290 kinases , identified in Table S1 ., An additional 5 uninformative kinases were removed from the panel as their percent inhibition values did not vary significantly across positive and negative SMKIs ., Thus , preprocessing yielded a data matrix of 113×285 for machine learning analysis ., A heatmap of the full dataset prior to preprocessing can be found in Figure S1 ., Of the 113 compounds , 30 and 83 SMKIs were classified as MNT positive and negative , respectively ., In the first phase of the analysis , several models were generated , each based on 10×5-fold cross validation for a particular combination of feature selection methods and a binary classifier ., In this phase , the best performing feature selection methods were a Kolmogorov-Smirnov/T-test hybrid algorithm , followed by Random Forests ., The most informative features were then input into a non-linear Support Vector Machines ( SVM ) classifier ., The Kolmogorov-Smirnov/T-test algorithm is a univariate filter method used to filter features based on their p-value ., Briefly , for each feature , the distribution of percent inhibition values was assessed ., If normal , a t-test was performed to yield a p-value ., Otherwise , a Kolmogorov-Smirnov test was run ., Features were ranked by p-value , and the top 100 features or less that met the 0 . 05 p-value cutoff were retained for further analysis ., The 100 or less features from the first method were then input into Random Forests 20 , a multivariate feature selection method based on decision trees ., In this phase of the analysis , Random Forests was used to select 10 features for input into the binary classifier ., SVMs are widely used in bioinformatics and other applications of supervised learning ., SVMs are used to find a hyperplane that maximizes the margin between the two classes of compounds in n-dimensional space , where in this analysis , n corresponds to the number of features selected using Random Forests ., Initial classification was performed using a nonlinear Radial Basis Function ( RBF ) kernel , with a cost of 1 and a gamma of 0 ., After selecting the model methods , the second phase of the analysis involved optimizing the model parameters ., The 10 splits of 5-fold cross validation were then used with the model methods to sweep over the number of features from 2 to 50 ., Optimal performance was achieved with 45 features ., To avoid overfitting and make the model generalizable to future compounds , we selected the minimum number of features whose model yielded an accuracy within one standard deviation of the performance obtained when using 45 features ., Thus we selected 21 as the number of features to select for the final model ., Using the full dataset , SVM cost and gamma parameters were then tuned ., Briefly , gamma is a parameter that affects the size of the hyperplane in an SVM , while cost is a penalizing measure for having a sample on the wrong side of the hyperplane ., Tuning yielded an optimal cost of 2 and a gamma of 2−4 using an RBF kernel ., Final model performance was based on a re-split of the data into 50 random splits of 10-fold cross validation ., Using the final model methods and optimized parameters , the 500 iterations yielded a cross-validated estimate with an overall classification accuracy of 85% ( standard deviation 1 . 8% ) , sensitivity of 68% ( standard deviation 5 . 0% ) , and a specificity of 91% ( standard deviation 2 . 0% ) ., A receiver operating characteristic ( ROC ) curve of the cross-validated and overall mean performance is shown in Figure 2 ., From the 50 splits of 10-fold cross validation used to assess final model performance , the frequency that each feature was selected as significant in Random Forests was tabulated ., The features were then ranked , and the top 21 most frequently-selected kinases were chosen as the model kinase profile ., The 21 model kinases are CAMK1 ( NP_003647 . 1 ) , CAMK2A ( NP_741960 . 1 ) , CAMK2D ( AAD20442 . 1 ) , DYRK1B ( NP_004705 . 1 ) , MAPK15 ( NP_620590 . 2 ) , PCTK1 ( NP_006192 . 1 ) , PCTK2 ( CAA47004 . 1 ) , PCTK3 ( NP_002587 . 2 ) , PFTK1 ( NP_036527 . 1 ) , CDK2 ( NP_001789 . 2 ) , CDK3 ( NP_001249 . 1 ) , CDK5 ( NP_004926 . 1 ) , GSK3A ( NP_063937 . 2 ) , CLK2 ( NP_003984 . 2 ) , MELK ( NP_055606 . 1 ) , BRSK2 ( NP_003948 . 2 ) , STK3 ( NP_006272 . 2 ) , MYLK ( NP_444254 . 3 ) , FLT3 ( NP_004110 . 2 ) , EIF2AK2 ( NP_002750 . 1 ) , and PRKAA2 ( NP_006243 . 2 ) ., Table 1 lists the 21 kinases and the frequency that each was selected as significant in this phase of the analysis ., A heatmap of the percent inhibition values against these 21 kinases is shown in Figure 3 and demonstrates a clear enrichment of kinase inhibition for SMKIs with MNT positive results ., To verify the statistical significance of the kinases , a dropout experiment was run using the preprocessed dataset , minus the 21 model kinases ., Using the same methods and 50 splits of 10-fold cross validation , the performance of the modified dataset ( 113 compounds×264 kinases ) was assessed ., The dropout model yielded an accuracy of 78% ( standard deviation 2 . 4% ) , sensitivity of 54% ( standard deviation 5 . 9% ) , and specificity of 87% ( standard deviation 2 . 0% ) ., All performance metrics for the dropout model yielded values at least one standard deviation worse than the original model , demonstrating the significance of the 21 identified kinases ., Additionally , to address multiple comparisons concerns , the q-values were calculated for all model kinases ., The FDR , or False Discovery Rate 21 , estimates the expected proportion of false positives in the data , and in this case , was based on the p-values derived from the Kolmogorov-Smirnov/T-test algorithm on all 290 kinases across all 113 compounds ., Q-values , which represent the minimum FDR at which each feature may be called significant , were then calculated using the “qvalue” package in R 22 ., At an FDR of 0 . 05 , 73 kinases of 290 may be called significant , including all model kinases ., At an FDR of 0 . 01 , 21 kinases may be called significant , although this includes only 10 of the model kinases ., Q-values for all model kinases are listed in Table 1 ., This result is expected since the kinases were selected based on both the filtering of feature selection one and the multivariate criterion of FS2 ., To verify the biological significance of the kinases , a review of literature was performed to find studies that might prove or suggest a mechanistic link between the model kinases and mitosis or genetic toxicological damage ., The majority of the kinases selected by this analysis ( 12/21 ) are members of the CMGC kinase family which is known to be involved with the control of cell proliferation ., The cyclin dependent kinases ( CDKs ) are a family of CMGC kinases that have been associated with mitosis and the cell cycle , and generally speaking , bind with cyclins during the various phases of mitosis ., While the three CDKs ( 2 , 3 , & 5 ) in the model all appear to have inhibition that is specific to the MNT positive compounds ( Figure 3 ) and are members of a family of kinases known to have roles in mitosis , only CDK2 has supporting literature making it biologically relevant to chromosomal damage ., Little is known about the cellular function of the other CMGC members of the CDK kinase family , such as PFTAIRE ( PFTK1 ) and PCTAIREs 1–3 ( PCTKs ) ., From this family , PFTK1 is the only kinase with literature supporting its biological relevance ., PFTK1 is a CDK2-related protein kinase which has been reported to phosphorylate the tumor suppressor Rb and interact with p21 , suggesting that PFTK1 is involved in cell cycle regulation 23 ., The activity of PCTK1 is cell-cycle dependent and displays a peak in the S and G2 phases 24 ., Selected kinases in the second largest group ( 7/21 ) are members of the calmodulin mediated kinase ( CAMK ) family , of which only MYLK has been reported to interact with chromosomes ., The smooth muscle myosin light chain kinase ( smMLCK or MYLK ) , which facilitates the movement of anaphase chromosomes through its involvement with actin and myosin 25 , is the sole kinase from this family with reported association with chromosome kinetics or the cell cycle , which has also been reported to induce spindle disruptions leading to metaphase arrest and chromosome defects 26 ., We applied a statistical modeling framework to identify a panel of kinases that are predictive of a positive micronucleus test result , a sign of potential chromosomal damage ., To our knowledge , this approach is the first application of a computational method to correlate high-throughput kinase screening results with a toxicological endpoint ., The described mathematical model is capable of predicting MNT results correctly 85% of the time based solely on compound inhibition profiles against 21 kinases ., The model presented herein indicates that chromosomal damage induced by many of the tested small molecule kinase inhibitors ( SMKIs ) correlates to their kinase inhibition profiles and that this knowledge can be used to design compounds with improved safety profiles at earlier stages of drug discovery ., While the 21 kinases identified in this analysis are statistically significant for our given dataset , our understanding of their mechanistic roles in chromosomal segregation and mitosis is still in its early stages ., A heatmap of the inhibition values against these 21 kinases ( Figure 3 ) displays a general pattern: SMKIs that are MNT negative tend to inhibit the 21 kinases much less frequently than the SMKIs that are MNT positive ., While some kinases selected by the model have a known function in cell cycle or chromosomal segregation , others have an unrelated or unknown role ., The majority of the kinases chosen are members of the CMGC kinase family , which is known to be involved with the control of cell proliferation ., While the role of CDK2 is well-documented , the biological relevance of CDK3 and CDK5 is less clear ., CDK3 was identified to complex with cyclin C and phosphorlyate the retinoblastoma tumor suppressor protein 27 and mediates the G1-S transition of the cell cycle 28 ., However , there are no reports of its involvement with chromosomal segregation ., CDK5 has been described as an unusual member of the CDK family because it has little known role in cellular proliferation and is activated by non-cyclin proteins 29 ., The three CDKs ( 2 , 3 , & 5 ) chosen all appear to have inhibition that is specific to the MNT positive compounds ( Figure 3 ) , and are members of a family of kinases known to have roles in mitosis , thus it is plausible that their inhibition could cause aberrant mitosis and errors in chromosomal segregation ., Other model kinases selected from the CMGC family are not known to have a clear role in chromosomal segregation ., Collectively , it appears that inhibition of the PCTK kinases ( PCTAIREs 1–3 ) strongly associates with chromosomal damage though this information has not been previously published ., The family of MAPKs is well-known to have a fundamental role in mitosis and cell cycle control 30 , and are involved with chromosome damage and micronucleus formation 31 , though this is the first report placing MAPK15 in such a role ., Glycogen synthase kinase 3 alpha ( GSK3a ) has been reported to be involved with chromosome alignment and cytokinesis 32–35 , thus its selection by mathematical modeling is not unanticipated ., However , GSK3a is one of the more promiscuous kinases selected in the mathematical model , with a large number of micronucleus negative compounds also inhibiting the kinase above 80% at the 10 µM concentration ( Figure 3 ) ., These data contrast with reported literature , as they suggest that combinations of kinase inhibition , rather than just GSKa alone , may be required for chromosome damage ., Model kinases in the calmodulin mediated kinase ( CAMK ) family were the most specific with regard to their inhibition of micronucleus positive compounds ( Figure 3 ) and were repeatedly chosen by the model , yet the majority , including CAMK1a , CAMK2a and 2d , MELK , BRSK2 , and PRKAA2 , do not have any reported function involving mitotic chromosome dynamics ., The two remaining kinases are FLT3 and MST1 , neither of which have any known role in either chromosomal segregation or mitosis ., FLT3 has been reported to associate with acute lyphoblastic leukemia 36 and chromosomal instability in the form of hyperdiploid aneuploidy observed in the same disease 37 , but its inhibition has not been linked to the missegregation of chromosomes ., MST1 , also known as STK3 , has been identified to be a substrate of caspases and play a role in apoptosis 38 ., Besides lack of validation of the mechanistic relevance of some model kinases , additional experimental limitations are that kinase-independent mechanisms of micronucleus formation exist while others are based upon analysis parameters ., In any analysis , the robustness of modeling complex endpoints is limited by the training dataset ., Care was taken in selecting which compounds to include and were chosen by judiciously sampling our internal kinase inhibitor library for SMKIs representative of broad scaffolds and designed for a variety of kinase targets ., Nevertheless , data were limited to past and current kinase projects at Roche and may not reflect future efforts ., Another potential concern with this approach is that not all kinases are available in the Ambit competition binding assay ., At the time of the analysis , only 290 of the 518 protein kinases were tested ., While the 290 kinases account for a large portion of the kinome and do not appear to miss large branches of the kinome tree ( data not shown ) , current panels are more comprehensive and will likely be more complete in the future ., In addition to the inherent limitations of the training set , the modeling approach does not necessarily identify all kinases that are highly predictive of chromosome damage ., As an example , several kinases were statistically significant in FS1 but were not selected by the FS methods ., This is a challenge often observed while analyzing large data sets 39–41: features that correlate individually with the endpoint are not chosen because they may be correlated with other features that are more highly correlated with the endpoint , and thus these features become redundant ., Said differently , it is often the case that there is more than one set of features that is highly predictive of the endpoint ., Such observations have been made frequently in other areas of biological research where the number of features outnumbers the sample number , e . g . microarray studies 39–41 ., Despite limitations , the strengths of this method lie in its utility ., At Roche and other pharmaceutical companies , SMKIs are designed to inhibit a kinase that has a known role in the pathway or disease of concern ., While hundreds of compounds may be developed that strongly bind to their target kinase , it is not often clear whether inhibiting other , non-targeted kinases will affect the success of the compound in further stages of the pipeline , especially in toxicological studies , where many compounds often fail ., In the process of building the model , we found that promiscuous SMKIs often tested positive for micronuclei formation ( Figure S1 ) , and this was greatly enriched through model development ( Figure 3 ) ., While general promiscuity may be a relatively good marker for determining the outcome of a micronucleus assay , there are examples where it isnt ( Figure S1 ) , suggesting that specific inhibition of particular kinases is of relevance ., Providing information to medicinal chemists early in the lead identification/optimization process , beyond just general guidance of promiscuity , is critical to the success of such a strategy , as it provides direction for development as well as compound prioritization for additional development ., Knowledge of these 21 kinases within Roche has been of assistance in designing SMKIs where a decrease in genotoxicity has been observed for this compound class ., While the framework presented here provides a robust method for identifying kinases correlated to genotoxicity , causality must be addressed by other means , along with concerns about indirect mechanisms of action and kinases not included in the dataset ., There are several kinases that are not solely inhibited by MNT positive compounds , including CLK2 , FLT3 , GSK3a , MAPK15 , PCTK1 , and EIF2AK2 ., This raises the question of whether their inhibition specifically causes the micronucleus formation or if their inhibition requires the inhibition of other kinases for the induction of chromosome damage ., Because there is high sequence homology amongst kinases in the ATP binding pocket , it is possible that some of the selected kinases do not cause chromosomal damage , but instead are correlated with the inhibition of others that do ., Alternatively , there a number of MNT positive compounds that do not inhibit any of the kinases chosen in the model ., Possible explanations include: first , a number of mechanisms independent of kinase inhibition can influence mitotic chromosome dynamics and second , kinases , which have not been screened in this study , may influence the outcome of the micronucleus result ., Development of SMKIs that carry a genotoxic liability can occur , though the generation of additional data demonstrating that the mechanism of action occurs in a non-DNA reactive , threshold-observable manner , is often necessary to appease regulatory agencies ., These additional developmental complexities are best avoided , as they are expensive , time consuming , and no guarantee exists that attrition due to chromosomal damage will be avoided ., Thus , many companies prefer to spend time and resources during lead optimization to identify compounds free of such liabilities , rather than risk failure due to later-stage attrition ., The use of mathematical modeling to better understand what underlies such toxicities is one of the first steps in designing drugs free of these particular liabilities ., Additional studies can shed light on the underlying pathways possibly connecting the model kinases , as it is clear that all are involved in some larger biological network and should not be considered as independent features ., Experimental studies can help to confirm possible connections ., As a basis for such future investigations , our methodology provides a starting point for biological hypothesis generation , in addition to its utility as a computational model for predicting genotoxicity ., The 113 compounds used in this project were synthesized internally or purchased from Sigma Chemical company ( St . Louis , MO ) ., The structural diversity of the 113 compound training set was assessed by representing each compound with Extended Connectivity Fingerprints ( ECFP ) in Pipeline Pilot 6 . 0 42 , a molecular characterization of compounds as a 2-dimensional fingerprint ., ECFP for each compound was used as input for principal components analysis ( PCA ) , as shown in Figure 1 ., All 113 compounds were sent to Ambit Biosciences ( San Diego , CA ) for kinase selectivity analysis against 317 kinases using KINOMEscan assays ., These 317 kinases cover a large and diverse portion of the human kinome ., For each kinase in this high-throughput competition binding screen , ligand-bound kinase quantities are measured in the presence and absence of the compound ., Input values for this project are reported in terms of percent inhibition ( % ) for each compound against each of the 317 kinases ., These measurements provide a means for identifying on-target and off-target kinases , as well as for quantifying the selectivity or promiscuity of an SMKI ., This is performed in a cell free binding assay which is used as a surrogate for cellular kinase inhibition , which can be influenced by physical-chemical properties ( solubility and permeability ) that may impact intracellular concentrations and kinase inhibition ., The in vitro micronucleus assay was conducted according to a previously published protocol 43 ., Briefly , the established permanent mouse lymphoma cell line L5178Y tk+/− ( ATCC CRL 9518 ) growing in suspension was obtained from Covance Laboratories Ltd . ( Harrogate , UK ) ., The top dose for evaluation was generally selected to observe acceptable toxicity ( decrease of the relative cell count ( RCC ) below 50% ) or clear signs of precipitation in the aqueous medium ., Micronucleus results obtained when the RCC falls below 40% are not interpreted as this exceeds the cytotoxicity cut-off ., Soluble and non-toxic compounds are evaluated up to a maximal dose level of 5000 µg/mL or 10 mM whichever is lower ., The cell cultures were exposed to the test compound for 24 h and harvested either immediately or following a 24 h recovery period in case of cell cycle arrest ., For assessment of cytotoxicity cell numbers are scored at harvest with the use of a Coulter Counter and relative cell counts ( RCC , as % negative control ) were calculated ( population doublings and cell morphology were assessed in parallel ) ., 1000 cells per dose were scored with a magnification of 1000× and micronuclei were evaluated according to previously described criteria 44 ., A compound is considered to induce a significant level of micronuclei , and thus yield a positive MNT result , if one or more concentrations show at least a 2% frequency of micronucleated cells in either of the two testing regimens ( generally corresponding to a 2 . 5 fold increase over historical controls ) ., Methylmethanesulfonate ( 15 µg/ml ) is used as a micronucleus positive control ., MNT assay results were ultimately considered binary , with a positive result corresponding to toxicity and a negative result to non-toxicity ., While some compounds were easily classified , others required reclassification because of experimental parameters ., Because Ambit data was measured at a 10 µM concentration for all SMKIs , positive micronucleus results that occurred above this 10 µM cutoff could be due to kinase inhibition that would not be reflected in the kinase inhibition assay results ., Thus , to better correlate the kinase inhibition to micronucleus positive results , compounds identified to be micronucleus positive above this threshold were reclassified as negative ., The study was based on a training set of 113 internal SMKIs ., To make the model generalizable to the prediction of future compounds , a large and structurally dissimilar group of SMKIs was selected ., The training set compounds were chosen to represent a diversity of molecular structures , physicochemical properties , and kinase targets ., Each compound was assessed for chromosomal damage , a sign of potential toxicity , using an in vitro micronucleus test ., Additionally , each compound was screened in a competition binding assay to quantify inhibition of 317 kinases ., Preprocessing was performed prior to employing machine learning methods ., From the 317 kinase panel , mutant kinases were removed from the dataset as their mechanistic function would be difficult to interpret ., Additional kinases were removed because their percent inhibition level , usually less than 50% at 10 uM , did not differ among the 113 compounds , making them uninformative when separating positive from negative MNT results ., When building a model to predict a binary endpoint , best practices for machine learning recommend using feature selection methods to reduce dimensionality of the data , followed by input into a pattern recognition method ., From observation , we have seen that while a selection of certain methods performs well with a given dataset , others methods do not ., Similarly , while there are many instances of machine learning methods that perform well , their performance results are not always reproducible with other datasets ., In this analysis we present a framework that involves sweeping over variety of methods , which allows the choice of methods to be driven by the data rather than investigator preference , and selecting methods and method parameter based on top-performing results ., An overview of the framework is given in Figure 4 ., Implementation details of the framework , such as choice of programming language and which methods to include , are left to the investigator ., This analysis was performed in R version 2 . 6 . 2 45 , based on the number of machine learning packages readily available ., The first phase of the analysis aimed to identify the machine learning methods to be used in the model ( Figure 4a ) ., We started by creating 10 random splits of 5-fold stratified cross validation ., Briefly , each split randomly grouped the 113 SMKIs into 5 subsets , or folds ., Each fold was stratified , meaning that the proportion of MNT positive compounds to MNT negative compounds in each fold roughly reflected that of the full dataset ., For each of the k folds , k-1 subsets were used as the training set to build the model , while the remaining subset was used as the test set to estimate performance ., For each fold within each split of data , a combination of two feature selection ( FS ) methods were run , followed by a binary classifier ., FS methods were used to determine which kinases , or features , are likely to correlate most with MNT result ., Feature selection methods were separated into two groups: univariate filter methods capable of handling larger input data ( FS1 ) , and more computationally-intensive multivariate methods ( FS2 ) ., FS1 methods consider features independently and are thus less likely to overfit to the given dataset ., Such methods include a Kolmogorov-Smirnov/T-test filter , single train error 46 , and ReliefF 47 ., However , since the FS1 methods do not address redundancy of features , multivariate approaches were employed in FS2 to consider correlation among a given subset of features ., FS2 methods included random forests 20 , genetic algorithm 48 , simulated annealing 48 , Gram-Schmidt Orthogonalization 49 , and RFE-SVM 50 ., The in

Introduction, Results, Discussion, Materials and Methods

Kinases are heavily pursued pharmaceutical targets because of their mechanistic role in many diseases ., Small molecule kinase inhibitors ( SMKIs ) are a compound class that includes marketed drugs and compounds in various stages of drug development ., While effective , many SMKIs have been associated with toxicity including chromosomal damage ., Screening for kinase-mediated toxicity as early as possible is crucial , as is a better understanding of how off-target kinase inhibition may give rise to chromosomal damage ., To that end , we employed a competitive binding assay and an analytical method to predict the toxicity of SMKIs ., Specifically , we developed a model based on the binding affinity of SMKIs to a panel of kinases to predict whether a compound tests positive for chromosome damage ., As training data , we used the binding affinity of 113 SMKIs against a representative subset of all kinases ( 290 kinases ) , yielding a 113×290 data matrix ., Additionally , these 113 SMKIs were tested for genotoxicity in an in vitro micronucleus test ( MNT ) ., Among a variety of models from our analytical toolbox , we selected using cross-validation a combination of feature selection and pattern recognition techniques: Kolmogorov-Smirnov/T-test hybrid as a univariate filter , followed by Random Forests for feature selection and Support Vector Machines ( SVM ) for pattern recognition ., Feature selection identified 21 kinases predictive of MNT ., Using the corresponding binding affinities , the SVM could accurately predict MNT results with 85% accuracy ( 68% sensitivity , 91% specificity ) ., This indicates that kinase inhibition profiles are predictive of SMKI genotoxicity ., While in vitro testing is required for regulatory review , our analysis identified a fast and cost-efficient method for screening out compounds earlier in drug development ., Equally important , by identifying a panel of kinases predictive of genotoxicity , we provide medicinal chemists a set of kinases to avoid when designing compounds , thereby providing a basis for rational drug design away from genotoxicity .

Small molecule kinase inhibitors ( SMKIs ) are a class of chemicals that have successfully been used for the treatment of a number of oncological diseases that are now being pursued by the pharmaceutical industry for inflammatory diseases , such as rheumatoid arthritis ., SMKIs are generally designed to specifically inhibit one kinase , but this is challenging due to the structural similarity of the ATP binding pocket amongst different members of the kinase family ., The inability to selectively inhibit just one kinase can be problematic , as kinases play key roles in a number of cellular processes ., Thus the unwanted inhibition of additional kinases can lead to undesirable toxicities that may halt drug development ., One type of toxicity often observed with this class of compounds is damage to chromosomes , which can occur when kinases involved with cell cycle progression or chromosome dynamics are inhibited ., Here we demonstrate that mathematical modeling can be used to identify kinases that correlate with chromosome damage , information which can assist medicinal chemists in avoiding certain kinases when synthesizing new chemicals ., Generation of this type of information is one of the first steps in beginning to reduce toxicity-based attrition for this class of compounds .

mathematics/statistics, molecular biology, genetics and genomics/chromosome biology

journal.pbio.2006660

Direct visualization of single-molecule membrane protein interactions in living cells

Membrane proteins play crucial roles in communication between intracellular and extracellular environments across cell membranes 1 ., Malfunctioning of membrane proteins often results in myriad diseases 2 , which makes these proteins major therapeutic targets 3 ., Despite their importance in cell signaling and drug development , however , membrane protein interactions in living cells have been poorly understood due to methodological limitations 4 ., Various methods to investigate membrane protein interactions have been developed over several decades , such as chemical cross-linking , yeast two-hybrid ( Y2H ) , and fluorescence resonance energy transfer ( FRET ) 5 , 6 ., Nevertheless , the intrinsic principles of these assays are actually the same: proximity between a bait protein ( protein of interest ) and a prey protein ( binding partner ) is utilized for the measurement of their interaction ., The use of proximity between two proteins as an indirect indicator for their physical interaction can produce false positives , especially when the interactions in a crowded membrane are investigated 7 ., Furthermore , the readout signals of these assays rely on the distance between the tags of a bait and a prey , which varies the results depending on the tag orientation on the proteins and makes it difficult to directly and quantitatively translate the result into the strength of the interaction 8 , 9 ., The dimerization of receptors in a plasma membrane is a critical process for receptor activation 10 ., Although the structural aspect of receptor dimerization has been intensively studied 11 , 12 , information about the dynamics of the dimerization in a plasma membrane still remains elusive ., The characterization of transient dimerization under various conditions such as drug treatment or mutations is particularly difficult , mainly due to the limited ability of current tools to capture the rapid moment of the dynamic interaction in the crowded membrane of living cells 4 , 5 , 13 , 14 ., Here , we established an in situ imaging method that directly captures the membrane protein interactions in living cells on the basis of the protein’s inherent diffusivity by utilizing the synergy between single-particle tracking ( SPT ) and antibody-induced protein immobilization , of which powerfulness to assess the protein–protein interaction was previously demonstrated 15 ., The interaction between prey and bait proteins was visualized through the co-immobilization ( Co-II ) of the prey with the immobilized bait ., Then , the co-immobilizing event was counted at the single-molecule level using single-particle tracking photoactivated localization microscopy ( sptPALM ) 16 , allowing us to determine and compare the strength of the interactions in the membrane of living cells ., Using Co-II , we revealed that epidermal growth factor receptor ( EGFR ) and beta-2 adrenergic receptor ( β2-AR ) homodimerization are dominantly regulated by the intramolecular conformation and membrane microenvironment , respectively ., To directly visualize protein–protein interactions in the plasma membrane of living cells at the single-molecule level , a bait protein ( a protein of interest ) on a cell membrane is specifically immobilized using its antibody coated on a glass surface ., Then , a prey protein ( an interacting partner ) that diffuses on the plasma membrane is immobilized together with the bait protein whenever the interaction occurs , which provides a direct indicator of their physical interactions ( Fig 1A ) ., This co-immobilized moment of the prey protein with the bait protein is captured by sptPALM 16 ., By counting the number of co-immobilized single-molecule trajectories specifically generated by the prey–bait interaction after the immobilization of the bait protein , the strength of the interactions can be quantitatively determined , allowing linear comparisons between the two interactions ., We call this method co-immunoimmobilization ( Co-II ) ., Co-II overcomes the limitations derived from the use of proximity , including false positives at high density , dependency on tag orientation , and difficulty of quantification ( Fig 1B ) ., Complete immobilization of bait proteins is critical for Co-II implementation; otherwise , the interactions between the prey and bait proteins do not always produce co-immobilized trajectories ., We examined the efficiency of the immunoimmobilization using EGFR ., To build an antibody-coated coverslip , we prepared a thiol-functionalized coverslip using 3-mercaptopropyl-trimethoxysilane ., Next , we utilized maleimide-activated neutravidin to covalently passivate the neutravidin to the coverslip and then added the biotin-conjugated antibody ., Using COS7 cells transiently expressing EGFR tagged with monomeric Eos fluorescent protein variant 3 . 2 ( mEos3 . 2 ) at its C terminus ( EGFR-mEos3 . 2 ) , we analyzed the immobilized fraction of EGFR , which increased after the addition of the anti-EGFR antibody using sptPALM ., To quantify the amount of the immobilized fraction , we calculated short-time diffusion coefficients from the trajectories to define immobilization in terms of diffusivity using mean squared displacement ( MSD ) = 4DΔt + 4e2 ( 0 < Δt < 780 ms ) ., The diffusion coefficient criteria for classifying immobilization were determined based on a localization error ., Nearly complete EGFR immobilization ( >93 . 3% ) was achieved when the secondary antibody was adopted between the neutravidin and anti-EGFR antibody to adjust the height between a glass surface and a plasma membrane ( S1 Fig ) ., The immobilization efficiency was independent of the expression level or the binding epitopes of EGFR targeted by different antibodies ( S1 Fig ) ., The even immobilization of the bait proteins was achieved across the entire cell surface within 15 min at 100 μg/mL of the antibody ( S2 Fig ) ., Another major concern for Co-II implementation was whether the immobilization of EGFR is specific to all the membrane proteins coexisting in a plasma membrane; otherwise , the co-immobilization of a prey protein would result from the interaction with nonspecifically immobilized proteins , not only with the intended bait protein ., We verified that the immobilization of EGFR using the anti-EGFR antibody coated on a glass surface did not immobilize various membrane proteins , including erb-b2 receptor tyrosine kinase 2 ( ErbB2 ) , erb-b2 receptor tyrosine kinase 3 ( ErbB3 ) , insulin receptor , β2-AR , and plasma membrane targeting ( PMT ) signal peptide , which force mEos3 . 2 to localize on a plasma membrane ., ( Fig 1C ) ., The immobilization of EGFR did not alter the spatial organization of EGFR distribution on the plasma membrane ( S2 Fig ) ., Furthermore , no cross-linking of the bait proteins induced by the anti-bait antibody was observed , as the excess amount of the antibody compared with the bait protein was coated on the glass surface ( S2 Fig ) ., To further evaluate the specificity of the immunoimmobilization , we simultaneously monitored the EGFR and β2-AR trajectories in a single cell using mEos3 . 2 and a SNAP tag labeled with benzyl-guanine–conjugated CF660R , respectively , before and after the addition of the anti-EGFR antibody ( Fig 1D ) ., When EGFR was immobilized with 98 . 1% of an immobilized fraction , the immobilized β2-AR fraction was not altered significantly ( S3 Fig ) ., This specificity of the immobilization between EGFR and β2-AR was confirmed vice versa ( Fig 1E and S3 Fig ) ., These results showed that Co-II can be simply and robustly implemented and should provide a direct indicator of the protein–protein interactions in the plasma membrane of living cells ., Using Co-II , we quantitatively measured EGFR pre-homodimerization ( ligand-independent dimerization ) in a live COS7 cell by utilizing EGFR-mEos3 . 2 as a prey and SNAP-EGFR as a bait ( Fig 2A ) ., To minimize the dimerization events between two mobile EGFR-mEos3 . 2 proteins , we excessively expressed SNAP-EGFR compared with EGFR-mEos3 . 2 , which allows us to assume the dimerization process as a pseudo-first-order reaction for the determination of an equilibrium dissociation constant , KD ( See the detail in Materials and methods ) ., We tracked CF660R-labeled SNAP-EGFR and EGFR-mEos3 . 2 before and after the addition of the anti-SNAP antibody ( Fig 2B ) ., The mobile subpopulation of SNAP-EGFR was almost fully shifted into the immobile subpopulation ( 95 . 2% ) , whereas only a partial shift ( 22 . 7% ) was observed for EGFR-mEos3 . 2 ( Fig 2C and 2D ) , which represents the amount of the physical interaction between the mobile EGFR-mEos3 . 2 and the immobilized SNAP-EGFR at dynamic equilibrium ., We also observed the transient colocalization of the prey EGFR with the immobilized bait EGFR at the single-molecule level , which supports that the co-immobilization of the prey EGFR is derived from a physically interacting process ( S4 Fig ) ., Next , we determined the concentration of the immobilized bait EGFR in the plasma membrane , which is determined by the concentration of SNAP-EGFR multiplied by the anti-SNAP antibody-induced immobilization fraction of SNAP-EGFR ., The concentration of SNAP-EGFR on the COS7 cell surface was measured by normalizing the total fluorescence intensity by the single-molecule intensity 17–19 ( Fig 2E ) ., We obtained a total internal reflection fluorescence ( TIRF ) image for CF660R-labeled SNAP-EGFR prior to the tracking procedure ., After the tracking procedures were finished , we acquired TIRF images for single-molecule SNAP-EGFRs by photobleaching until individual SNAP proteins were spatially resolved ., We collected single-molecule SNAP-EGFRs that exhibited a one-step photobleaching trace to calculate the average fluorescent intensity emitted from a single CF660R dye ., We additionally corrected the total number of SNAP-EGFRs , considering the proportion of nonfluorescent CF660R , which should be immobilized but not detected 20 ( S5 Fig ) ., Because membrane proteins diffuse laterally on a two-dimensional plasma membrane , we used a density notation instead of molar concentration because the definition of molarity in a plasma membrane is currently ambiguous 18 , 21 , 22 ., We assumed that the plasma membrane is flat because the in situ measurement of the actual geometry of the dynamic plasma membrane is technically currently limited 18 , which may cause bias in the estimation of the concentration 21 ., We analyzed the dependency of the co-immobilized fraction of EGFR-mEos3 . 2 with respect to the expression level of SNAP-EGFR ( Fig 2F ) ., The KD of EGFR pre-homodimerization in the single cell was 973 ± 47 molecules/μm2 ( mean ± SEM ) in Dulbeccos Modified Eagle Medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) at 37 °C ., Analysis using a Scatchard plot confirmed that Co-II measured the pseudo-first-order reaction of the EGFR pre-dimerization ( Fig 2F ) ., This result indicates that the major portion of the interaction between the mobile prey and the immobilized bait is a dimerization process at the single-cell level in the concentration range we measured ., The KD values of the EGFR pre-dimerization in various cell lines , including HeLa , HEK293 , and CHO-K1 cells , did not vary significantly ( Fig 2G ) , indicating that the contexts of plasma membrane from these cell lines marginally affect the EGFR pre-dimerization ., No photodamage was detected in the cells after the Co-II assay ( S6 Fig ) ., The in situ capability of Co-II allows us to acquire a spatial KD map in a single living cell ., The power of sptPALM to obtain the sufficient number of trajectories in a single cell enables us to determine the KD in a small area of the single cell ., We constructed a KD map of EGFR pre-homodimerization in a single living cell at a 1 . 2-μm resolution ( Fig 2H ) ., The distribution of the KD values obtained from different regions of plasma membrane showed a log-normal distribution with a KD of 1 , 059 ± 612 molecules/μm2 ( mean ± SD ) ., We found a geometric tendency of KD values in the plasma membrane to be lower at the periphery and higher at the center ( Fig 2I and 2J and S7 Fig ) , which is consistent with the previous report regarding the spatial control of EGFR activation 23 ., Although this spatial heterogeneity of KD values might result from the bias in receptor concentration derived from the assumption of a flat membrane , this spatial heterogeneity implies that the intrinsic characteristics of EGFR pre-homodimerization might be controlled by the cellular microenvironment in living cells ., The ability of the in vivo KD measurement using Co-II led us to explore the simple but currently unresolved question of how much the KD values of EGFR dimerization decrease upon epidermal growth factor ( EGF ) stimulation ., Above all , we confirmed that the Co-II system did not perturb ligand-induced receptor activation ( S8 Fig ) ., The KD values of EGFR homodimerization determined by Co-II were 122 ± 14 and 1 , 606 ± 332 molecules/μm2 with and without EGF , respectively , in DMEM without serum at 37 °C ( Fig 3A and S9 Fig ) ., The decrease in KD produced by EGF was approximately 13 . 2-fold ., The effects of nonnatural EGFR ligands on EGFR dimerization were further examined ., First , we measured the KD of EGFR dimerization in the presence of the fragment antigen-binding ( Fab ) of cetuximab , which blocks the extended conformation of EGFR extracellular domain ( ECD ) 24 ., As a result , the Fab fragment of cetuximab substantially impaired EGFR dimerization with and without EGF ( 610 ± 86 and 9 , 507 ± 5 , 450 molecules/μm2 , respectively ) ( Fig 3B ) ., The incomplete inhibition of EGF-induced EGFR dimerization by the Fab fragment might be due to the reduced affinity of the Fab fragment toward EGFR ( about 2 nM ) , which is an order of magnitude lower than that of EGF for high-affinity binding ( <0 . 2 nM ) 24 , 25 ., We also measured the KD of EGFR dimerization after treatment with erlotinib or lapatinib , which target an ATP binding pocket in the intracellular domain ( ICD ) of EGFR 26 ( Fig 3B and S10 Fig ) ., Erlotinib reduced the KD of EGFR dimerization without EGF , while lapatinib exerted an insignificant effect ., By contrast , the inhibition potency of EGF-induced EGFR dimerization was significantly higher in lapatinib ., These results indicate that erlotinib and lapatinib have different preferences on the active and inactive EGFR conformations , consistent with recent molecular structures revealed by cryo-electron microscopy ( cryo-EM ) 27 ., Next , we explored the effects of intramolecular changes on the KD of EGFR dimerization utilizing two EGFR mutants frequently found in various cancers , EGFRvIII and EGFR L858R 28 ., The KD values for EGFRvIII and EGFR L858R dimerization were significantly decreased by an average of approximately 6 . 5- and approximately 2 . 4-fold compared with that for EGFR WT , respectively ( Fig 3C and S10 Fig ) ., These decreased KD extents indicate that these oncogenic mutants form a substantial level of dimers at their physiological expression levels in cancer , consistent with previous reports that their ligand-independent activity is derived by enhanced dimerization in cancer 29 , 30 ., A scale mapping the KD values of EGFR pre-homodimerization from the various inter- and intramolecular perturbations to EGFR was drawn ( Fig 3D ) ., Interestingly , the KD values induced by perturbations to EGFR ECD tended to span a much broader range than that to EGFR ICD ., Unliganded ECD conformation of EGFR has been previously controversial 11 , 18 , 30 , so the contribution of EGFR ECD to EGFR pre-homodimerization was unclear ., Our quantitative comparison of the KD values provides direct evidence that EGFR ECD contributes more critically to EGFR pre-homodimerization than EGFR ICD does , which is consistent with a recent report displaying the dynamic conformational changes of unliganded EGFR ECD using solid-state NMR 31 ., Recently , the β2-AR homodimer was probed using proximity-based methods , including bioluminescence resonance energy transfer ( BRET ) , although its existence still has been controversial because of methodological concerns 32 , 33 ., Using Co-II , we determined the KD of the β2-AR pre-homodimerization in a live COS7 cell without serum ( 1 , 508 ± 145 molecules/μm2 ) , which suggests the existence of a mixture of both β2-AR monomer and homodimer at a typical physiological expression level in living cells 34 , although the KD value measured in situ using Co-II is about 3-fold lower than the value measured in vitro using proteoliposome , likely due to the effect of the microenvironmental context of a plasma membrane 35 , 36 ., Interestingly , the KD of the β2-AR pre-homodimerization is similar to that of EGFR pre-homodimerization ( 1 , 606 ± 332 molecules/μm2 ) , which led us to further investigate the differences between the homodimers of these two receptors in distinct receptor classes ., The addition of isoproterenol decreased the KD of the β2-AR dimerization by about 3 . 3-fold ( 462 ± 82 molecules/μm2 ) , unlike EGF , which decreased the KD of the EGFR dimerization by about 13 . 2-fold ( 122 ± 14 molecules/μm2 ) ., This result is possibly derived from the lack of an explicit structural interface for the β2-AR dimerization such as the dimerization arm of EGFR extended by EGF 11 , 37 ., Both EGFR and β2-AR have been previously reported to be regulated by membrane microenvironment , such as a cholesterol 38 , 39 ., We compared the KD values of the homodimerizations of the two receptors after sequestrating cholesterol in a plasma membrane ., Surprisingly , the dimerization of β2-AR was markedly disrupted by nystatin ( 22 , 378 ± 4 , 283 molecules/μm2 ) , whereas that of EGFR was significantly enhanced ( 453 ± 95 molecules/μm2 ) , indicating that EGFR and β2-AR homodimerizations are differentially regulated by the membrane microenvironment ., A scale mapping the KD values of the homodimerizations of these two receptors under ligand treatment and cholesterol depletion was drawn in Fig 4 ., Co-II analyzes membrane protein interactions based on their inherent diffusivity instead of their proximity , which is utilized for prevalent methods ., This principle of Co-II liberates concentration dependency , which is critical when proximity is used as an indicator for the physical interaction , because random collision between noninteracting proteins can frequently occur at high concentration ., Co-II provides reliable data even at the high density in a crowded membrane of living cells , as no interaction between EGFR and PMT was observed even at a saturated expression level ( KD = 26 , 890 ± 2 , 724 molecules/μm2 ) ( S11 Fig ) ., Furthermore , Co-II is conceptually independent of the tag orientation on the proteins because the intrinsic property of a protein itself is the subject of the measurement in Co-II , whereas the tag is the subject in the proximity-based methods 8 ( S10 Fig ) ., Therefore , the bona fide analysis using Co-II could provide unprecedented quantitative information regarding membrane protein interactions affected by natural ligands , drugs , mutations , and microenvironmental changes in a single living cell ( Figs 3 and 4 ) ., Although single-molecule trajectories contain convoluted information regarding multiple molecular processes , the interpretation of protein diffusion has been subjective; the changes of diffusion coefficient were interpreted as one distinct molecular process based on a theoretical assumption , without thorough experimental verification 23 , 40 ., Furthermore , transient colocalization among single-molecule trajectories has been presumed as their molecular interaction , even though the colocalization is only a necessary condition for the physical interaction because the localization accuracy of fluorescent proteins is not sufficient to resolve direct molecular interactions 41 , 42 ., Although two-color quantum dot tracking has circumvented this problem by observing colocalizing trajectories with a correlated motion , the probability of physical interaction between two sparsely visualized fluorescently labeled proteins is extremely low , which restrains this approach from directly assessing the number of interacting molecules to determine equilibrium constants 43 ., Thus , the lacking objectiveness for biologically interpreting trajectory data has limited the application of SPT to specific research and requires elaborate experimental controls to prevent the misinterpretation of data ., These problems are mainly derived from the fact that the natural change in the diffusion coefficient made by the interaction between two diffusing proteins is marginal ., However , the objective deconvolution of interaction information from single-molecule trajectories becomes possible with Co-II , because the change in diffusion coefficient by the interaction is represented by an order of magnitude difference in the diffusion coefficient ( from about 0 . 2 μm2/s to about 0 . 008 μm2/s ) and the change appears by a controllable trigger , the antibody addition ., The application of Co-II gets retarded as the diffusion of a prey is slowed , due to the classification error between mobile and immobile trajectories ., The diffusion coefficient of a prey up to 0 . 04 μm2/s can be applied for Co-II using mEos3 . 2 , considering the full width at half maximum of the log distributions of EGFR and β2-AR diffusion coefficients obtained by using mEos3 . 2 ., Compared with previous methods to detect protein–protein interactions by using protein immobilization and the bulk measurement of fluorescent intensity 44 , there are several major advantages in Co-II ., First , it enables the direct visualization of single-protein interactions in living cells , which makes it possible to perform single-molecule research in living cells ., Because membrane proteins in the plasma membrane of living cells coexist at very high concentration and constantly flow , the spatiotemporal positions of individual proteins cannot be accurately determined , even using super-resolution microscopy ., Co-II overcame this concentration problem in the membrane of living cells by utilizing the dimension of a protein’s intrinsic diffusivity in addition to space and time dimensions ., Second , high measurement sensitivity is achieved from the large number of single-molecule data ., It was possible to precisely probe about 3% of the co-immobilized EGFR fraction using more than 10 , 000 single-molecule trajectories ., This high sensitivity cannot be reached by the bulk fluorescence intensity , which fluctuates at high level and is vulnerable to photobleaching ., This sensitivity issue becomes critical as the protein–protein interaction of interest is weaker or more transient ., Third , quantitative information is derived from the counting of single molecules , enabling robust and precise quantification with a linear dynamic range ., Furthermore , Co-II does not suffer from photobleaching because diffusivity , not fluorescent intensity , is the measurement , which generates reliable data even with multiple measurements ., Lastly , it might provide relative stoichiometry information ., We analyzed the frequency of stopping EGFR in the vicinity of the immobilized one , which enables us to infer the distribution of the oligomer size ( S12 Fig ) ., This stoichiometry analysis implies that EGFR dimer is a major population , with a small portion of oligomers induced by EGF ., More extensive experiments might be required to verify whether the recently reported EGFR tetramers exist at a significant level 45 , 46 ., KD values over a wide range , encompassing both strong and weak interactions , can be analyzed using Co-II simply by controlling the expression of bait proteins ., Eq 3 in the Materials and methods section provides the optimal expression range of a bait protein for determining KD ( S13 Fig ) ., The interactions more than 1 , 000 times stronger or 10 times weaker than EGFR pre-homodimerization can be resolved ., The measurement of high KD values for weak interactions becomes possible due to the ability to capture the rapid transient interactions between membrane proteins using SPT 47 ., Because Co-II utilizes the immobilization of the one of the reactants , the measured reaction rate for homodimerization should be equal to the true rate divided by two , according to the Smoluchowski reaction rate , if the reaction of interest is diffusion controlled 48 ., Although biochemical reaction kinetics on the plasma membrane might be affected by the crowdedness or the microdomains of the plasma membrane , which contribute to proteins’ diffusivity , it is not clear whether receptor dimerization is actually diffusion controlled ., In case of EGFR , the conformational change of EGFR ECD from a tethered form to an extended form is crucial for its dimerization 49 , implying that the activation energy is a major factor for EGFR dimerization ., Furthermore , EGF binding marginally affects the diffusion coefficient of EGFR , according to the Saffman-Delbruck model 50 and our measurement ., Care must be taken to interpret KD values measured by Co-II , considering whether the reaction of interest is governed by activation energy or diffusion ., Conversely , this bias might be useful to characterize whether the reaction process is activation controlled or diffusion controlled ., Co-II should not be limited to statistically analyzing KD at the ensemble level ., The power of Co-II can be expanded to provide dynamic interaction constants such as a dissociation constant ( koff ) and an association constant ( kon ) at the single-molecule level and reveal the single-molecule heterogeneity of membrane protein interactions in living cells ., By obtaining long trajectories using a photoswitchable organic dye , Alexa Fluor 647 , we directly visualized the dissociation process ( the mobile-immobile-mobile transition ) of single-molecule EGFR pre-homodimerization ( S4 Fig ) ., Although the probability of observing the process was substantially low ( 0 . 001 ) due to the insufficient duration of trajectories obtained by Alexa Fluor 647 , the measured koff value ( about 1 . 2 s−1 ) was similar to the previous report measured by the two-color colocalization of quantum dot trajectories 43 ., Repetitive interactions of a single mobile prey with immobilized baits can be observed if a fluorescent probe or a nanoparticle that yields sufficiently long trajectories is utilized ., Recently , the distinct regulation of ErbB3 phosphorylation by the interaction with EGFR upon the stimuli of different ligands was reported 51 , in which HER3 dimerization and clustering with EGFR are differentially controlled by different ligands ., Along with this finding , our result that EGFR and β2-AR homodimerizations are differentially regulated by cholesterol demonstrates that the microevironment of the plasma membrane is critically involved in their activation mechanism in living cells , which lies on shared context with previous reports 52 , 53 , 54 ., These observations together strongly suggest that receptor activation is differentially regulated by both the intramolecular conformation and the microenvironment of the plasma membrane in living cells ., Co-II should be useful to elucidating the dynamic changes of membrane protein interactions in the diverse physiological contexts of living cells and understanding the precise regulation of receptor activation in the membranes of living cells ., To construct the mEos3 . 2 fusion protein at the N terminus of EGFR , we subcloned human EGFR into the pcDNA3 . 1 vector ( V800-20 , Invitrogen ) with the following primers 1–4 ., Then , mEos3 . 2 extracted from pEGFP-N1/mEos3 . 2 , a kind gift from Dr . Tao Xu ( Chinese Academy of Science ) , was inserted between the signal and mature peptide of EGFR with the following primers 5–6 ., To construct SNAP-tagged EGFR , the SNAP tag gene from the pSNAPf vector ( N9183S , New England Biolabs ) was subcloned into pcDNA3 . 1/mEos3 . 2-EGFR with the following primers 7–8 ., The SNAP-tagged EGFRvIII ( SNAP-EGFRvIII ) and EGFR L858R constructs ( SNAP-EGFR L858R ) were obtained by replacing the EGFR WT gene from pcDNA3 . 1/SNAP tag-EGFR with the EGFRvIII and EGFR L858R genes using the following primers 9–10 and 11–12 , respectively ., To construct the mEos3 . 2-tagged InsR at the C terminus , we first subcloned the InsR gene , a kind gift from Ingo Leibiger ( Karolinska Institutet , Sweden ) , into pcDNA3 . 1/mEos3 . 2-His at the N terminus of mEos3 . 2 with the following primers 13–14 ., To construct SNAP-tagged β2-AR , we subcloned the SNAP tag gene into the N terminus of β2-AR with the signal peptide from hemagglutinin to enhance membrane localization ., The corresponding templates were obtained from Matthew Meyerson ( Addgene plasmid #11011 for EGFR WT; Addgene plasmid #11012 for EGFR L858R ) , Alonzo Ross ( Addgene plasmid #20737 for EGFRvIII ) , and Robert Lefkowitz ( Addgene plasmid #14697 for β2-AR ) ., All the other plasmids , including PMT-mEos3 . 2 , EGFR-mEos3 . 2 , EGFRvIII-mEos3 . 2 , EGFR L858R-mEos3 . 2 , ErbB2-mEos3 . 2 , ErbB3-mEos3 . 2 , and β2-AR-mEos3 . 2 , were prepared as previously described 47 ., Primer 1: 5′-CGCAAATGGGCGGTAGGCGTG Primer 2: 5′-CCGCGGTTGGCGCGCCAGCCCGACTCGCCGGGCAGAG Primer 3: 5′-GGCGCGCCAACCGCGGCTGGAGGAAAAGAAAGTTTGC Primer 4: 5′-AGCTTTGTTTAAACTTATGCTCCAATAAATTCACTGCT Primer 5: 5′-GGCGCGCCACATCATCACCATCACCATATGAGTGCGATTAAGCCAGAC Primer 6: 5′-TCCCCGCGGCCCTCCACTCCCACTTCGTCTGGCATTGTCAGGCAA Primer 7: 5′-GGCGCGCCACATCATCACCATCACCATATGGACAAAGACTGCGAAATG Primer 8: 5′-TCCCCGCGGCCCTCCACTCCCACT ACCCAGCCCAGGCTTGCCCAG Primer 9: 5′-TCCCCGCGGCTGGAGGAAAAGAAAGGTAAT Primer 10: 5′-AGCTTTGTTTAAACTCATGCTCCAATAAATTCACT Primer 11: 5′-TCCCCGCGGCTGGAGGAAAAGAAAGTTTGC Primer 12: 5′-AGCTTTGTTTAAACTCATGCTCCAATAAATTCACT Primer 13: 5′-CGGGATCCATGGCCACCGGGGGCCGGCGG Primer 14: 5′-GCTCTAGAACTCCCGGAAGGATTGGACCGAGGCAA The antibodies and reagents were obtained from the following vendors: the mAb 199 . 12 ( AHR5072 ) and Alexa Fluor 647–conjugated anti-mouse antibody ( A21235 ) were obtained from Invitrogen; both mAb 528 ( sc-120 ) and mAb R-1 ( sc-101 ) were obtained from Santa Cruz; the SNAP tag antibody ( CAB4255 ) , rabbit anti-mouse IgG ( 31194 ) , biotin-conjugated EGFR antibody ( MA5-12872 ) , and anti-6x His tag antibody ( MA1-21315 ) were obtained from Thermo Scientific; the anti-mEos3 . 2 antibody ( A010-mEOS ) was purchased from Badrilla; the anti-phosphorylated EGFR antibody ( Y1068 , ab32430 ) was obtained from Abcam; the anti-actin antibody ( 691001 ) was obtained from MP Biomedicals; cetuximab was obtained from Merck Serono; erlotinib and lapatinib were obtained from Selleckchem; and EGF ( E9644 ) , nystatin ( N6261 ) , and isoproterenol ( I5627 ) were purchased from Sigma-Aldrich ., The cetuximab Fab fragment was generated from an intact antibody using a Fab preparation kit ( 44685 , Pierce ) , and cetuximab was labeled with Alexa 647 dye using the Alexa Fluor 647 Antibody Labeling Kit ( A20186 , Thermo Scientific ) ., CF660R , succinimidyl ester ( 92134 , Biotium ) , was reacted with BG-NH2 ( S9148S , New England Biolabs ) in dimethylformamide while shaking at 30 °C overnight according to the manufacturer’s instructions ., The solvent was vacuum-evaporated and the product was dissolved in distilled water after purification by HPLC ., COS7 , HEK293 , and HeLa cells were obtained from American Type Culture Collection ( ATCC ) and cultured in DMEM ( Lonza ) supplemented with 10% FBS ( Gibco ) at 37 °C , 5% CO2 , and 95% humidity ., CHO-K1 cells ( ATCC ) were cultured in DMEM/F-12 1:1 modified medium ( Thermo Scientific ) supplemented with 10% FBS at 37 °C , 5% CO2 , and 95% humidity ., The cells were transfected using lipofectamine LTX ( Invitrogen ) according to the manufacturer’s instructions ., Glass coverslips were washed in chloroform/methanol ( 50/50 ) for 24 h and stored in ethanol ., After drying , the coverslips were oxidized in a plasma chamber ( Femto Science ) for 5 min and then incubated in a closed jar containing a silanization solution

Introduction, Results, Discussion, Materials and methods

Interactions between membrane proteins are poorly understood despite their importance in cell signaling and drug development ., Here , we present a co-immunoimmobilization assay ( Co-II ) enabling the direct observation of membrane protein interactions in single living cells that overcomes the limitations of currently prevalent proximity-based indirect methods ., Using Co-II , we investigated the transient homodimerizations of epidermal growth factor receptor ( EGFR ) and beta-2 adrenergic receptor ( β2-AR ) in living cells , revealing the differential regulation of these receptors’ dimerizations by molecular conformations and microenvironment in a plasma membrane ., Co-II should provide a simple , rapid , and robust platform for visualizing both weak and strong protein interactions in the plasma membrane of living cells .

Protein–protein interactions govern cellular processes ., The majority of these physical interactions previously identified are strong/permanent interactions , which typically remain unbroken even after purification ., The weak/transient interactions between proteins have been implicated in the control of dynamic cellular process that maintain cellular homeostasis and trigger signaling cascades upon environmental changes ., However , these interactions are poorly investigated , mainly due to the methodological limitations ., Here , we have developed a co-immunoimmobilization assay called Co-II that enables the direct visualization of protein–protein interactions in the membrane of living cells at the single-molecule level ., Co-II is based on the intuitive concept that if the protein of interest is immobilized , the interacting protein must be co-immobilized ., The use of intrinsic protein diffusivity fundamentally overcomes the limitations of proximity-based methods ., Using Co-II , we study the transient homodimerizations of EGFR and β2-AR in living cells , which have been implicated in several types of cancers and heart diseases ., We show that the dimerization of these receptors is differently regulated by molecular conformations and the microenvironment in the plasma membrane .

fluorescence imaging, amorphous solids, medicine and health sciences, methods and resources, protein interactions, glass, endocrine physiology, membrane proteins, fluorophotometry, materials science, growth factors, cellular structures and organelles, epidermal growth factor, physical chemistry, chemical properties, research and analysis methods, fluorescence resonance energy transfer, imaging techniques, dimerization, proteins, endocrinology, chemistry, cell membranes, spectrophotometry, biochemistry, cell biology, physiology, biology and life sciences, physical sciences, materials, spectrum analysis techniques

journal.ppat.1000569

Plasmodium falciparum Heterochromatin Protein 1 Marks Genomic Loci Linked to Phenotypic Variation of Exported Virulence Factors

Plasmodium falciparum causes the most severe form of malaria in humans with over one million deaths annually 1 ., Severe and fatal outcomes of infections with this protozoan parasite result from a multitude of syndromes triggered by repeated rounds of asexual reproduction within erythrocytes ., After invasion into red blood cells ( RBCs ) , the parasite initiates a dramatic host cell remodeling process , culminating in the export of parasite virulence factors onto the surface of infected RBCs ( iRBCs ) 2 ., The majority of these proteins is encoded by species-specific subtelomeric gene families , some of which underwent massive expansion during the evolution of the P . falciparum lineage 3 ., One of the direct consequences of their concerted expression is the sequestration of iRBCs in the microvasculatory system , a process that is linked to severe complications including cerebral and placental malaria 4–6 ., Sequestration occurs due to interactions of P . falciparum erythrocyte membrane protein 1 ( PfEMP1 ) with various receptors on endothelial cells and uninfected erythrocytes 7–10 ., 60 PfEMP1 variants are encoded by individual members of the var gene family 11–13 ., Importantly , only one var gene is transcribed by a single parasite ( mutual exclusion ) 14 ., Switches in var gene transcription occur in situ in absence of any apparent recombination events and result in antigenic variation of PfEMP1 15 ., This clonal phenotypic variation allows the parasite to evade variant-specific humoral immune responses and to sequester in various tissues 16 ., Members of some other gene families ( rif , stevor , Pfmc-2tm ) are also expressed in a restrictive manner 17–19 , however , their role in parasite biology and the underlying regulatory mechanisms remain unclear ., Multiple var genes are located in most subtelomeric regions directly downstream of telomere-associated repeat elements ( TAREs ) and in internal clusters on some chromosomes 20 ., At either location var genes occur in close association with other variably expressed multi-gene families 13 ., Several recent studies investigating the nature of the epigenetic mechanisms involved in the control of mono-allelic var gene transcription revealed an important role of the conserved promoter and intron sequences 21 ., var promoters are silenced by default and , notably , activation of an episomal var promoters caused silencing of the entire repertoire of native var genes 22–24 ., In addition to the 5′ upstream sequences the var gene intron is involved in silencing 25 and , although its exact role in this process remains controversial , this finding has been confirmed several times 26–28 ., Fluorescent in situ hybridisation ( FISH ) experiments revealed that P . falciparum chromosome ends occur in perinuclear clusters 29 , 30 ., Consequently , subtelomeric var genes are inherently positioned at the nuclear periphery which is linked to enhanced transcriptional silencing in other eukaryotes 31–33 ., Moreover , this spatial association was also demonstrated for chromosome-internal var genes 23 , 34 ., Transgenes inserted into subtelomeric repeat regions were transcriptionally silenced in a metastable fashion 35 , similar to position-effect variegation in yeasts and higher eukaryotes 36 ., The process of var gene activation thus appears to be linked to nuclear re-positioning of a silenced locus into a transcriptionally active zone lending support for the existence of a specialized perinuclear region dedicated to mutually exclusive var transcription 23 , 35 , 37 ., The perinuclear location of var genes is clearly independent of their transcriptional state , however , conflicting results exist as to whether var gene activation occurs within or outside chromosome end clusters 23 , 30 , 34 , 35 , 37 ., Together , this complex architectural setup provides a dynamic foundation for the heritable and variegated silencing of var genes and other variably expressed gene families by epigenetic control processes ., The current data is consistent with facultative heterochromatin-based silencing where dynamic alterations in local chromatin structure act as the major regulatory mechanism ., Many of the reversible histone modifications characteristic for active or silenced chromatin in other eukaryotes have been described in P . falciparum 38–40 ., Recent studies addressed their role in var gene regulation and uncovered the first indications for a distinct histone code linked to variegated var gene expression 41 ., Active var genes are associated with histone 3 acetylation ( H3K9ac ) and methylation ( H3K4me2 , H3K4me3 ) and silenced var genes are enriched in H3K9me3 39 , 42 ., The P . falciparum genome also encodes a core set of histone-modifying enzymes including histone deacetylases and methyltransferases ( HMTs ) 43 , 44 ., Interestingly , different subsets of var and rif genes show increased transcription in mutant parasite lines lacking either one of the two PfSIR2 paralogs , demonstrating a direct role for these histone deacetylases in virulence gene silencing 35 , 45 ., Recently , H3K9me3 has been mapped to the full set of var genes and to members of other subtelomeric gene families on a genome-wide scale 37 , 46 ., The presence of H3K9me3 at silenced var loci provides important clues about the possible control strategy underlying epigenetic var regulation ., This particular histone modification is a conserved hallmark of heterochromatic silencing and serves as a docking site for heterochromatin protein 1 ( HP1 ) to nucleate the propagation of heterochromatin along the chromosome fiber 47 , 48 ., Indeed , PfHP1 , the P . falciparum ortholog of HP1 , binds to H3K9me3 and was shown to be involved in variegated expression of a particular var gene 49 ., Although these findings justify the assumption that PfHP1 may be associated with H3K9me3-enriched loci , the genome-wide localization of PfHP1 is unknown and cannot be inferred directly from these data; HP1 proteins interact with a multitude of other proteins 50 , proteins other than HP1 bind to H3K9me3 51 , H3K9me3 is not necessarily sufficient to recruit HP1 52 , 53 , and HP1 can be recruited to heterochromatin in an H3K9me3-independent manner 54 , 55 ., In this study , we provide a comprehensive analysis of PfHP1 on the level of protein function and genome-wide distribution ., We show that PfHP1 binds specifically to H3K9me3 but not to other repressive histone methylation marks ., Using nuclear fractionation and detailed immuno-localization experiments , we demonstrate that PfHP1 constitutes a major heterochromatin component with confined localization to perinuclear foci ., High-resolution analysis by genome-wide chromatin immunoprecipitation ( ChIP-on-chip ) revealed a striking PfHP1 occupancy pattern restricted to 425 genes , most of which are members of P . falciparum-specific exported virulence families ., The majority of these genes were also enriched in H3K9me3 underscoring the biological significance of this interaction in virulence gene expression ., In addition , we detected 38 PfHP1-bound genes not enriched in H3K9me3 ., Many of these genes code for invasion proteins or proteins specifically expressed in different life-cycle stages suggesting a previously unrecognized role for PfHP1 in invasion pathway switching and life cycle progression ., Furthermore , we show that PfHP1 is not associated with centromeric regions implying important differences in centromere biology compared to other eukaryotes ., Consistent with a role of PfHP1 in virulence gene silencing we find that over-expression of PfHP1 leads to downregulation of 78 genes , the majority of which are located in heterochromatic domains ., In summary , our results attribute an important role to PfHP1 in parasite biology and suggest a unifying PfHP1-dependent mechanism by which P . falciparum regulates the variegated expression of proteins involved in virulence and phenotypic variation ., HP1 belongs to a family of conserved chromatin proteins found from fission yeast to humans and is characterised by an N-terminal chromodomain , which binds H3K9me3 , and a C-terminal chromoshadow domain implicated in homo- and heterodimerisation 56 ., The P . falciparum genome encodes three putative chromodomain proteins amongst which PFL1005c was recently identified as the P . falciparum ortholog of HP1 ( PfHP1 ) 49 ., A multiple sequence alignment based on structural information on HP1 revealed a remarkable similarity between the chromodomains in PfHP1 and those in HP1 from various eukaryotes , including conservation of residues critical for interaction with H3K9me3 57 , 58 ( Figure S1 ) ., Similarly , most of the conserved residues important for the C-terminal chromoshadow domain and homo- and hetero-dimerisation are also present 57 , 59–61 ., To confirm that PfHP1 indeed displays these structural and functional HP1-like properties we expressed PfHP1 in E . coli ., Using pull down experiments , we showed that PfHP1-HIS binds specifically to H3K9me3 but not to unmethylated H3 ( Figure S2 ) ., Furthermore , we experimentally verified homo-dimerisation of PfHP1 by mixed incubation of PfHP1-HIS with nuclear extracts prepared from 3D7/HP1-Ty parasites expressing a 2×Ty-tagged version of PfHP1 ( Figure S2 ) ., These findings provide an independent confirmation of the results obtained by Perez-Toledo and co-workers 49 identifying PFL1005c as the P . falciparum ortholog of heterochromatin protein ., To further test the specificity of the PfHP1-H3K9me3 interaction , we investigated binding of PfHP1 to a set of alternative histone modifications ., Methylation of two other lysine residues in H3 and H4 , H3K27me3 and H4K20me3 , respectively , are commonly associated with transcriptional repression 62 , 63 ., Furthermore , phosphorylation of serine 10 adjacent to K9me3 ( H3K9me3S10p ) was found to counteract the repressive effect of H3K9me3 by preventing HP1 binding 64 , 65 ., Using a peptide competition assay , we show that PfHP1 interacts specifically with H3K9me3 , whereas H3K9me3S10p , H3K9ac and H4K20 peptides were unable to interact with PfHP1 ( Figure 1A ) ., H3K27me3 interfered weakly with PfHP1-binding to H3K9me3 which was also observed for mouse HP1β 58 ., The H3K9me3S10p and H3K27me3 marks have not been detected in P . falciparum to date 40 and therefore the biological significance of these findings remains to be determined ., However , it is tempting to speculate that P . falciparum may use phosphorylation of H3S10 to reverse H3K9me3-mediated repression ., Consistent with its localization to heterochromatic regions , PfHP1 was insoluble after serial extraction of isolated parasite nuclei with low and high salt buffers ( Figure 1B ) ., After complete digestion of native chromatin with micrococcal nuclease ( MNAse ) or DNAseI a substantial fraction of PfHP1 remained associated with the high salt-insoluble pellet ., Interestingly , the association of PfHP1 with the insoluble nuclear fraction was not sensitive to treatment with RNAse suggesting that in contrast to other eukaryotes , binding of PfHP1 to chromatin does not require RNA components 66–68 ., These results demonstrate a tight association of PfHP1 with highly compact heterochromatic structures and/or the nuclear matrix ., Since var genes are dynamically associated with chromosome end clusters , we expected PfHP1 to be located in such defined perinuclear regions ., To test this hypothesis we used three independent transgenic parasite lines ., 3D7/HP1-GFP expresses a PfHP1-GFP fusion protein from its endogenous promoter , and 3D7/HP1-HA and 3D7/HP1-Ty express epitope-tagged versions of PfHP1 from episomal plasmids ., Live cell imaging revealed that PfHP1-GFP located to the nucleus in a punctate perinuclear pattern reminiscent of chromosome end clusters ( Figure 2A ) ., 3D reconstruction verified the position of PfHP1 foci to the nuclear periphery ( Video S1 ) ., Indirect immunofluorescence assays ( IFA ) confirmed these results and accentuated the localization of PfHP1 to discrete and well-defined regions at the nuclear periphery ( Figures 2B and C ) ., On average we observed 3 . 6 PfHP1-HA signals per trophozoite stage nucleus ( 3 . 64 ( mean ) ±1 . 34 ( s . d . ) ; 302 nuclei counted ) ., A detailed IFA experiment confirmed this restricted localization pattern in parasites carrying a single or multiple nuclei throughout intra-erythrocytic development ( Figure S3 ) ., Next , we used immunoelectron microscopy to characterise PfHP1 localization at the ultrastructural level in 3D7/HP1-GFP parasites ., PfHP1 was dominantly located at the nuclear periphery within and adjacent to previously described electron-dense regions reflecting perinuclear heterochromatin ( Figure 2D ) 34 ., These high-resolution immunolocalizations corroborated and extended the light microscopy results , and together these data support the presence of PfHP1 at heterochromatic regions in the nuclear periphery ., The PfHP1 localization pattern is consistent with multiple punctate foci of HP1 observed in S . pombe , plant and mammalian nuclei 69–71 ., Finally , we asked if this discrete perinuclear localization pattern corresponds to chromosome end clusters by testing for a direct association of PfHP1 with subtelomeric DNA using combined IFA/FISH ., In greater than 80% of cases PfHP1-GFP signals occurred directly adjacent to , or overlapped with , the subtelomeric repeat probe rep20 suggesting that PfHP1 is a major component of telomeric clusters ( Figure 3 ) ., Surprisingly , we obtained a higher number of signals per nucleus for rep20 ( 5 . 02 ( mean ) ±1 . 66 ( s . d . ) ) compared to PfHP1 ( 3 . 3±1 . 21 ) ., In summary , our microscopy-based data identify PfHP1 as a major component of chromosome end clusters suggesting a preferential association of PfHP1 with P . falciparum subtelomeric regions ., In light of the above results and the recently published genome-wide H3K9me3 patterns 37 , 46 it was tempting to speculate that PfHP1 plays a dominant role in subtelomeric virulence gene silencing ., To test this hypothesis and to identify additional potential PfHP1 target loci we performed genome-wide chromatin immunoprecipitation ( ChIP-on-chip ) using a high-density whole genome tiling array ( NimbleGen Systems Inc . ) 46 ., We observed a striking association of PfHP1 with subtelomeric regions on all parasite chromosomes ( Figure 4A , Tables S1 and S2 ) ., These domains covered TARE repeat blocks and extended inwards including all subtelomeric var genes ., Additional internal PfHP1-bound islands were identified on chromosomes 4 , 6 , 7 , 8 and 12 , which in most cases defined chromosome-central var gene clusters ., Hence , PfHP1 binds to the full complement of subtelomeric and chromosome-internal var genes ., PfHP1 occupancy was not restricted to var loci but instead covered extended regions including all members of other gene families shown to be expressed in a clonally variant manner , including rif , stevor and Pfmc-2tm 17–19 ( Figure 4B ) ., This peculiar association is even more striking considering that PfHP1 was hardly detected at loci falling outside these chromosomal regions ., In fact , the immediate boundaries between PfHP1-occupied and PfHP1-free regions delineate the sites of species-specific indels or synteny breakpoints between P . falciparum and other Plasmodium species 3 , 72 ., 95% of all 425 PfHP1-bound genes code for P . falciparum-specific proteins most of which are involved in host-parasite interactions ( Figures 4C , S4 and Table S2 ) ., Almost all of these genes are members of subtelomeric gene families coding for proteins exported to the erythrocyte ( var , rif , stevor , pfmc-2tm , surfin , pfacs , fikk kinases ) or predicted to be exported ( phista , phistb , phistc , dnajI , dnajIII , a/b hydrolases , hyp1 to hyp17 ) ( Figure 4C and Table 1 ) ., In addition , some other lineage-specific genes were also occupied by PfHP1: invasion-related genes Pfrh1 73 and Pfrh3 ( pseudogene ) 74 , liver stage antigen 1 ( lsa1 ) 75 , the gametocyte-specific gene Pf11-1 76 , non-syntenic tRNA and rRNA loci , and a number of genes coding for hypothetical proteins ., Surprisingly few PfHP1-bound loci code for proteins with orthologs in other Plasmodium species ., These include members of the rhoph1/clag family ( clag2 , clag3 . 1 , clag3 . 2 ) involved in erythrocyte invasion 77; genes implicated in the development of sexual stages: ccp1 78 , 79 , pfs230 80 , 81 , putative dynein heavy chains 82; crmp1 and crmp4 expressed in sporozoites 83; and PFL1085w , coding for an ApiAP2 transcription factor 84 ., Compared to Salcedo-Amaya and colleagues 46 , who used the same NimbleGen array for genome-wide H3K9me3 mapping , we found a high level of local as well as genome wide correlation between the presence of this repressive mark and PfHP1 ( R2\u200a=\u200a0 . 72 ) ( Figure 4A and B ) ., Out of 425 PfHP1-bound genes , we found only 44 genes that were not classified as H3K9me3-enriched , and one H3K9me3-associated gene was devoid of PfHP1-binding ., In contrast , 155 PfHP1-associated genes were not classified as H3K9me3-enriched in the study by Lopez-Rubio et al . ( 68 of which were not represented on their array ) 37 , while all but six genes enriched in H3K9me3 were also occupied by PfHP1 ., By lowering the enrichment threshold for the latter study , PfHP1-occupancy correlated well with both genome-wide H3K9me3 localization datasets showing that 387 out of 425 PfHP1-bound loci were consistently enriched in H3K9me3 ( Table S2 ) ., Most of these genes are members of gene families coding for proteins exported to the erythrocyte and implicated in parasite virulence ( Figure 4C ) ., Surprisingly , about half of the genes bound by PfHP1 but devoid of the H3K9me3 mark are single copy genes and members of small gene families coding for invasion proteins or proteins expressed in different life-cycle stages ( Figure 4C and Table S2 ) ., This raises the interesting possibility that PfHP1 may be recruited to these loci in an H3K9me3-independent manner ., Alternatively , this discrepancy may be related to overexpression of PfHP1 , or to the use of different parasite lines and/or ChIP protocols ., To validate the ChIP-on-chip results we investigated the association of PfHP1 with individual loci by ChIP-qPCR ., We targeted ten and twelve randomly selected loci , which showed either a negative or a positive association with PfHP1 in the ChIP-on-chip experiment , respectively ( Figure 5 and Figure S5 ) ., These results confirmed the ChIP-on-chip findings in all instances , showing that PfHP1 is associated with subtelomeric and internal virulence genes but not with genes that showed no association with PfHP1 in the ChIP-on-chip experiment ( Figure 5A ) ., No chromatin fragments from transgenic parasite lines were recovered with rabbit IgG control antibodies or anti-HA/anti-GFP antibodies used on 3D7 wild-type parasites ( data not shown ) , demonstrating the specificity of these results ., Importantly , and consistent with a role of PfHP1 in stably inherited heterochromatic silencing , PfHP1-occupancy was present at the same loci in next generation ring stage parasites ( Figure 5B ) ., As expected , PfHP1-positive genes were also enriched in H3K9me3 , which confirms the genome-wide colocalization and underscores the in vivo relevance of the PfHP1/H3K9me3 interaction in virulence gene silencing and the mutually exclusive presence of H3K9ac and H3K9me3 ( Figure 5C ) ., In contrast , genes not bound by PfHP1 were generally enriched in H3K9ac although this association did not necessarily correlate with active transcription of these loci ., This is not surprising in light of recent findings demonstrating that H3K9ac occupancy did not differ markedly between the coding regions of active and inactive genes in P . falciparum 46 ., In summary , our results demonstrate an extraordinarily confined localization of PfHP1 throughout the genome , and an extensive colocalization with the repressive histone mark , H3K9me3 ., This implies an important role for PfHP1 in epigenetic regulation of exported virulence factors and indicates that variegated expression and phenotypic variation may represent a general , rather than exceptional , feature of most P . falciparum-specific or expanded gene families ., In other eukaryotes like S . pombe and D . melanogaster , HP1 is an important factor in centromere function and a major constituent of pericentromeric heterochromatin 48 ., We observed no evident presence of PfHP1 in these domains , albeit the average level of PfHP1 ChIP-on-chip signal over genes directly adjacent to centromeres was somewhat higher as compared to the rest of the genome ( Figure S6 and Table S1 ) ., To test if PfHP1 was indeed enriched at centromeres we performed ChIP-qPCR experiments targeting the centromeres on eight chromosomes and six genes directly up- or downstream of centromeres which displayed low-level PfHP1-binding in ChIP-on-chip ., We were unable to detect binding of PfHP1 to these regions as none of the loci tested showed any sign of PfHP1 enrichment ( Figure S6 ) ., These findings are in line with the observed absence of H3K9me3 marks at centromeric regions and suggest that P . falciparum centromere biology and chromosome segregation are independent of PfHP1 ., HP1 has been implicated in gene silencing 85 , 86 and hence we were interested in testing this proposed function of PfHP1 in P . falciparum ., Although recent whole transcriptome analyses strongly suggest that most subtelomeric gene families are expressed in a restricted manner 87 , 88 , a formal demonstration of a direct link between PfHP1 and gene expression is lacking ., We therefore focused on a possible genome-wide correlation by transcriptional profiling using RNA isolated at four consecutive timepoints across the intra-erythrocytic developmental cycle ( IDC ) from two biological replicates of the PfHP1-overexpressing line ( Table S3 ) ., PfHP1 target genes displayed significantly lower absolute expression levels as compared to all other genes at all IDC stages ( p<0 . 001 , Wilcoxon ranksum test ) ( Figure S7 ) ., It is noteworthy that many PfHP1-negative genes are also weakly or not expressed during the IDC , stressing the notion that PfHP1 is not a general marker for inactive genes and that other processes such as gene-specific regulation participate in developmental and cell-cycle-dependent transcriptional control ., We were also interested in testing the effect of perturbations in PfHP1 expression on global gene transcription ., Several attempts to generate a PfHP1-null mutant failed suggesting an essential role for this protein in parasite biology ., We therefore investigated if over-expression of PfHP1 had any effect on gene transcription by comparing mRNA levels of all genes in the transfected lines to those in a control line ., We detected 78 genes that were consistently down-regulated in two biological replicates and none that were upregulated ( Figure 6 ) ., Of these , 50 are members of PfHP1-demarcated gene families , 28 of which showed a greater than three-fold enrichment for PfHP1 in the ChIP-on-chip experiment ( p-value 7 . 76E-36 ) , including nine of the ten pfmc-2tm family members ., Importantly , this analysis identified additional PfHP1 target genes that were either not detected , or classified as below three-fold enriched , in the ChIP-on-chip experiment and showed no sign for H3K9me3 enrichment ., These include all members of the hyp5 family and additional members of the crmp , ccp , eba and dynein heavy chain families ., At this stage , however , it remains unknown if the down-regulation of these genes is due to a direct or indirect effect of PfHP1 over-expression ., Hence , increased levels of PfHP1 enhanced silencing of variegated genes and had only minor effects on global gene transcription ., These results are consistent with a dosage-dependent effect of PfHP1 and suggest further that unwanted heterochromatin spreading is efficiently prevented by defined boundary structures ., Furthermore , our approach demonstrates that over-expression studies by transcriptional profiling may be employed to investigate the function of regulatory proteins in P . falciparum gene expression ., In this study we present a comprehensive analysis of P . falciparum HP1 and describe the first genome-wide binding profile of a non-histone chromatin component in this important pathogen ., Our findings reveal important insights into the regulatory strategy employed to control the variegated expression of a large class of highly specialized virulence genes ., This knowledge will be instrumental for future investigations to understand parasite virulence and survival ., We have shown that PfHP1 binds specifically to H3K9me3 and forms stable homodimers in vitro , which are both conserved features of HP1 in other eukaryotes ., A recent study used similar approaches to demonstrate these biochemical features for PfHP1 49 ., In vivo , PfHP1 associates extensively with subtelomeric repeats and genes encoding virulence factors in both subtelomeric and chromosome-internal loci ., The conserved organisation of subtelomeric regions into blocks of distinct subtelomeric repeat units followed by multiple members of various gene families is a hallmark feature of P . falciparum chromosome ends ., Unknown protein ( s ) mediate physical linking of chromosome ends in the formation of telomeric clusters 30 ., The findings presented here identified PfHP1 as a major constituent of these chromosome end clusters ., Our nuclear fractionation results suggest that PfHP1 is a candidate protein responsible , at least in part , for the physical clustering of chromosome ends through interactions between the chromoshadow domain and other structural components ., Whether chromosome-internal heterochromatic domains are an integral part of chromosome end clusters , or rather represent physically distinct entities at the nuclear periphery remains a matter of debate ., Our ChIP-on-chip results demonstrated that PfHP1 associates with all subtelomeric regions ., Hence , if chromosome-internal PfHP1-enriched regions form entities distinct from telomeric clusters one would expect a higher number of perinuclear PfHP1 domains compared to the four to seven chromosome end clusters usually detected by FISH 29 , 30 , which we never observed ., Furthermore , in our IFA/FISH experiments the average number of rep20 signals was higher than that of PfHP1 ., At this stage , we dont know if this observation is due to the actual absence of PfHP1 from some chromosome end clusters or to differential sensitivities of IFA- and FISH-based target detection ., However , both results clearly argue against a location of central heterochromatic domains separate from telomeric clusters ., On the other hand , our IFA/FISH results may also be consistent with the idea that the few perinuclear PfHP1 foci not co-localising with rep20 reflect chromosome-internal heterochromatic regions that are physically distinct from chromosome end clusters , as has been suggested by others 37 ., We believe that both opposing hypotheses are related to technical limitations inherent to FISH and IFA/FISH experiments resulting in a failure to detect the full complement of DNA and protein targets simultaneously ., To know which scenario reflects the in vivo situation , more refined approaches such as locus tagging , confocal microscopy and/or 3C and 4C chromosome conformation capture techniques need to be applied ., We have shown that the majority of heterochromatic protein-coding genes are located subtelomerically directly adjacent and in smooth transition to the non-coding TARE region ., A number of additional PfHP1 islands are also found at chromosome-internal clusters ., In total , PfHP1 binds to 425 genes reflecting 7 . 5% of the parasites coding genome ., Notably , all heterochromatic coding domains are contained within sharply defined boundaries , which in most cases reflect non-syntenic regions ., In other words , nearly all of PfHP1-bound genes code for proteins that do not have orthologs in other organisms and are thus specific to the P . falciparum lineage ., This set of PfHP1-bound genes compares well with the genome-wide pool of H3K9me3-enriched loci described recently 37 , 46 ., This strong correlation is highly relevant for our understanding of the in vivo role of the PfHP1/H3K9me3 interaction and underscores its significance in P . falciparum virulence gene silencing ., Some genes associated with PfHP1 were enriched in H3K9me3 in the Salcedo-Amaya study 46 but not in the Lopez-Rubio study 37 ., These include members of the surfin , fikk kinase and gbph families as well as members of uncharacterised gene families such as hypxx and exported co-chaperones 3 ( see also Table 1 ) ., We attribute this discrepancy to an improved performance of native versus formaldehyde-crosslinked H3K9me3 ChIP rather than to the actual absence of this histone mark at these loci ., The vast majority of PfHP1-demarcated gene families code for proteins that are exported into the host erythrocyte to participate in the processes of host cell remodeling , immune evasion and cytoadherence 3 , 89 , 90 ., A hallmark in the epigenetic regulation of var gene transcription is the strict mutually exclusive expression of a single family member ., The demonstrated association of PfHP1/H3K9me3 with var genes significantly advances our knowledge of the mechanisms underlying mutually exclusive var gene expression and may serve as a model system to understand the regulation and biological role of other virulence gene families ., Clonal variation was also experimentally demonstrated for expression of a subset of PfHP1-bound gene families including rif , stevor , pfmc-2tm and surfin 17–19 , 91 ., Furthermore , several transcriptional profiling studies indicate restricted transcription of additional exported gene families that are also associated with PfHP1/H3K9me3 ., In view of the well-described role of HP1 in regional gene silencing , this remarkable association hints at an overall strategy to control phenotypic variation of a large pool of protein families that evolved to facilitate survival in a hostile environment ., The expansion of lineage-specific exported protein families is much more pronounced in P . falciparum compared to other Plasmodia 3 ., This observation is most likely related to the trafficking of PfEMP1 and other proteins to the erythrocyte surface and probably associated with the high virulence of P . falciparum ., It is therefore tempting to speculate that the continuous expansion of P . falciparum exported gene families from single ancestral gene types ultimately required the parallel evolution of an epigenetic system to ensure phenotypic variation and avoid premature exhaustion of the antigenic repertoire ., It is noteworthy that all but one ( PFI1780w , phistc ) of the genes coding for the core complement of 36 exported proteins shared between different Plasmodium species 3 are not associated with PfHP1 ., This is indicative for conserved essential functions of these ancestral proteins in the trafficking of exported proteins that evolved before lineage-specific expansion of virulence gene families ., The multi-step process of merozoite invasion into erythrocytes is characterised by the sequential action of apically located proteins encoded by gene families such as eba , Pfrh and rhopH/clag 92 ., Variations in the expression of these genes are linked to alternative invasion pathways involving different ligand-receptor interactions ., For instance , in isogenic 3D7 lines using either a sialic-acid dependent or independent invasion pathway , clag2 , clag3 . 1 and clag3 . 2 show clonal variation and , for the latter two genes , are transcribed in a mutually exclusive manner 93 ., Similarly , members of the Pfrh and eba families were shown to be differentially expressed in different parasite strains 94–97 ., The association

Introduction, Results, Discussion, Materials and Methods

Epigenetic processes are the main conductors of phenotypic variation in eukaryotes ., The malaria parasite Plasmodium falciparum employs antigenic variation of the major surface antigen PfEMP1 , encoded by 60 var genes , to evade acquired immune responses ., Antigenic variation of PfEMP1 occurs through in situ switches in mono-allelic var gene transcription , which is PfSIR2-dependent and associated with the presence of repressive H3K9me3 marks at silenced loci ., Here , we show that P . falciparum heterochromatin protein 1 ( PfHP1 ) binds specifically to H3K9me3 but not to other repressive histone methyl marks ., Based on nuclear fractionation and detailed immuno-localization assays , PfHP1 constitutes a major component of heterochromatin in perinuclear chromosome end clusters ., High-resolution genome-wide chromatin immuno-precipitation demonstrates the striking association of PfHP1 with virulence gene arrays in subtelomeric and chromosome-internal islands and a high correlation with previously mapped H3K9me3 marks ., These include not only var genes , but also the majority of P . falciparum lineage-specific gene families coding for exported proteins involved in host–parasite interactions ., In addition , we identified a number of PfHP1-bound genes that were not enriched in H3K9me3 , many of which code for proteins expressed during invasion or at different life cycle stages ., Interestingly , PfHP1 is absent from centromeric regions , implying important differences in centromere biology between P . falciparum and its human host ., Over-expression of PfHP1 results in an enhancement of variegated expression and highlights the presence of well-defined heterochromatic boundaries ., In summary , we identify PfHP1 as a major effector of virulence gene silencing and phenotypic variation ., Our results are instrumental for our understanding of this widely used survival strategy in unicellular pathogens .

Plasmodium falciparum causes the most severe form of malaria in humans ., The high virulence of this unicellular parasite is in part related to the selective expression of members of falciparum-specific gene families ., These genes encode proteins that are exported into the cytoplasm and onto the surface of infected red blood cells ., To avoid recognition by the hosts immune system , P . falciparum employs sequential expression of antigenically different variants of these surface proteins ., While the epigenetic mechanisms responsible for such clonal expression have been studied in some detail for the major virulence gene family var , the regulation and function of other exported protein families remain elusive ., Here , we identify P . falciparum heterochromatin protein 1 as a major structural component of virulence gene islands throughout the parasite genome ., This factor binds specifically to a reversible histone modification , which marks these virulence loci for transcriptional silencing ., Our observations suggest a unifying epigenetic strategy in the regulation of host–parasite interactions and immune evasion in P . falciparum ., Furthermore , these findings have important implications for the future study of hitherto uncharacterized exported proteins with roles in parasite virulence .

molecular biology/chromatin structure, molecular biology/histone modification, cell biology/gene expression, cell biology/nuclear structure and function, genetics and genomics/gene expression, genetics and genomics/nuclear structure and function, molecular biology/centromeres, genetics and genomics/chromosome biology, infectious diseases/protozoal infections, genetics and genomics/epigenetics, genetics and genomics/bioinformatics

journal.pcbi.1002046

Distributions of Transposable Elements Reveal Hazardous Zones in Mammalian Introns

Transposable Elements ( TEs ) are major factors that have shaped the landscape of the mammalian genome through evolution ., Most TEs in mammals are inactive remnants of ancient TE insertions , buried in the host genome for millions of years ., In rodents and primates , TEs comprise 38–45% of the genome 1 , 2 , and about 90% of all human RefSeq genes contain TEs in their introns ., These TEs can be divided into four major classes: long interspersed elements ( LINEs ) , short interspersed elements ( SINEs ) , long terminal repeat ( LTR ) retroelements ( including endogenous retroviruses ( ERVs ) ) , and DNA transposons 3 ., The first three classes are retrotransposons , which utilize an RNA intermediate during their retrotransposition process and account for most TEs in mammalian genomes ., On the other hand , DNA transposons move directly to new genomic loci without being reverse-transcribed ., Although most mammalian TEs are neutral components of the genome with no significant biological effects 4 , 5 , some elements do impact the cell/organism by acting as insertional mutagens , inducing DNA rearrangements , assuming cellular functions and altering gene regulation 4 , 6 , 7 , 8 ., Biologically significant TEs are usually discovered and studied on a case-by-case basis , although bioinformatics approaches have also been used to identify potentially functional TEs ., Genomic comparisons between species have identified deeply conserved TEs that function as regulatory elements 9 , ., TEs that serve as alternative exons , promoters or polyadenylation signals are also straightforward to detect by looking for chimeric transcripts between the TE and neighboring genes 11 , 12 , 13 , 14 ., Global TE distribution patterns in mammalian genomes have been intensely studied in the past decade , and such analyses have provided insight into the selective forces that influence fixation probabilities of TE insertions ., For example , some studies have evaluated the relationships between TE distributions and imprinted genes 15 , and gene expression patterns 16 , 17 , 18 ., TE-free regions have also been used as markers to identify potentially critical regulatory regions 19 , 20 ., Moreover , it is clear that LTR elements and LINEs are more prevalent in intergenic regions compared to gene introns , and most of those that do reside in gene introns are in the antisense orientation with respect to the enclosing genes 3 , 21 ., This pattern reflects stronger selection against sense-oriented elements , likely due to the greater chance that such elements will disrupt gene transcript processing 22 ., While cases have been reported of influential TEs far from genes , those elements near or within genes likely have a greater potential of affecting gene expression ., However , our current knowledge of the distribution of TEs within gene introns is very limited , and it remains unclear why some intronic TEs perturb gene transcription while most do not ., To fully understand their biological effects , it would be useful to determine which intronic TEs are most likely to affect gene expression , so they can be prioritized for functional analyses ., With a growing appreciation for SINE and LINE insertional polymorphisms in human 23 , 24 , 25 , 26 , 27 , 28 , such predictions would be particularly helpful in identifying polymorphic TE insertions with the greatest probability of affecting gene transcription and , therefore , possibly contributing to phenotypic variability or disease susceptibility in humans ., In this study , we conducted a set of bioinformatics analyses of TE distribution patterns within human and mouse genes and revealed TE underrepresentation zones and distributional biases in gene introns ., TEs that do occur within the underrepresentation zones are more likely to be involved in aberrant gene splicing and known cases of intronic disease-causing TE insertions are primarily located within these zones , strongly suggesting that TEs in these locations are more likely to be harmful and be selected against ., The results of our study reveal a distinct tendency for TEs to affect gene transcription when poised near exons , and point to their continued role in catalyzing genome evolution ., According to our genomic survey , 85–90% of mouse and human protein coding genes contain TE sequences in their introns ., In a recent study of the relationship between Alu SINEs and alternative splicing , Lev-Maor et al . reported a drop of Alu density within 150 bp from intron boundaries 29 ., Based on this observation and the fact that most intronic splice signals are located at the 5′- and 3′-end of introns 30 , we hypothesized that de novo intronic TE insertions near exons are more likely to be mutagenic , and consequently , that the frequency of TEs would be significantly lower than expected in general near intron ends ., To analyze the distributions of various TE classes within introns , we first conducted computer simulations to determine theoretical TE distribution patterns ( see Materials and Methods ) ., Then we determined the actual distribution pattern of intronic TEs according to their distance to the nearest exon ., To alleviate our concern about the potential effect of “distance shifting”- a hypothesized result of later TE insertions or other rearrangements occurring between a specific TE and its nearest exon , we also analyzed the distribution of the youngest 20% of intronic TEs ., However , we observed only minor differences compared to all intronic TEs in the genome ( data not shown ) ., To clearly show the difference between simulated and actual TE distributions at each predefined position in introns , we calculated the ‘standardized frequency’ of observed TEs ( see Materials and Methods ) ., Briefly , the level of TE representation at each predefined intronic interval is determined from the difference between the actual TE distribution in the genome ( observed ) and the computer simulation of random TE insertions ( expected ) ., When this value is positive , it reflects an overrepresentation of a given TE class within the corresponding intronic region; however , when negative it indicates underrepresentation ., As expected , we found that all four major TE classes are highly underrepresented near intron boundaries in both human ( Figure 1A in Text S1 ) and mouse ( data not shown ) ., We next applied the same distribution analysis for only full-length or near full-length TE sequences ( see Table 1 for “full-length” definitions ) ., Again , as shown in Figure 1B in Text S1 for human , full-length TEs were highly underrepresented when close to exons , but most TE classes except SINEs showed larger underrepresentation zones ( hereafter shortened to U-zone ) compared with the all-TE distributions ., We also noticed that intronic regions more than 20 kb from exons showed a significant underrepresentation of SINEs compared to random simulations ., Unlike patterns close to exons , intronic TE distributions greater than 20 kb from exons are less likely due to purifying selection so we searched for other explanations ., SINE elements are more abundant in G/C-rich regions 1 , 21 and , since large introns resemble intergenic regions in terms of G/C content ( which is generally A/T rich ) 31 , we postulated that the drop of SINE frequency compared to random simulations in deep intronic regions was an effect of local G/C content ., To determine if this was the case , we normalized our random simulations with the local G/C content as described in Materials and Methods ., Indeed , after applying such normalization , the underrepresentation of SINEs in deep intronic regions greatly flattened out , while the sizes of the U-zones near exons were not affected ., Hence all our subsequent analyses employed this normalization ., Figure 1 shows the normalized plots for all human TEs ( Figure 1A ) and full length TEs ( Figure 1B ) , and these plots are very similar for mouse TEs ( Figure 2 in Text S1 ) ., Interestingly , the sizes of the U-zones near intron boundaries are different between TE classes ( Table 1 ) ., Original insertion site preferences , natural selection and genetic drift could all contribute to global TE distributions ., While determining the initial integration site preference of TEs is difficult if not impossible ( especially for ancient families ) , a limited number of de novo TE integration studies showed that TEs in todays human genome are distributed very differently from their initial target site preferences 32 , 33 ., Indeed , since 99% of TEs in the human genome and 93% in the mouse genome have been fixed for more than 25 million years 1 , it is reasonable that their current distributions will bear little resemblance to any original insertion site preferences but will primarily be the result of selection and genetic drift ., Therefore , the TE U-zones identified here most likely result from purifying selection , rather than original avoidance of these regions during the integration process ., The larger U-zones for full length TEs ( compare Figures 1A and B ) suggests that purifying selection acts at much greater distances on full-length elements than on their partly deleted counterparts ., This effect is not observed for SINEs but these elements have a much shorter full-length size ( ∼300 bp for human Alus ) 1 , 8 , will generally carry fewer cryptic transcriptional regulatory signals and are less harmful to the enclosing genes than other TEs 34 ., For the above reasons , full-length SINE elements may be better tolerated at a closer distance to exons ., We next compared the average length of intronic TEs within and outside their full-length U-zones and found a significant difference for all TE classes in both species ( Figure 2 for human; Figure 3 in Text S1 for mouse ) ., In fact , most elements within their respective U-zones are truncated , while a greater portion of TEs beyond such zones are full-size elements , resulting in a much bigger size variance ( see the difference between upper whiskers in Figure 2 for human and also Figure 3 in Text S1 for mouse ) ., Therefore , the length of individual TEs is an important aspect dictating their genomic distributions , indicating that larger elements are more likely to be genotoxic when positioned near exons ., These results also support previous work regarding L1 LINEs , indicating that , compared to shorter elements , full length L1s have more potentially disruptive splice and polyadenylation signals 35 , have greater effects on expression of enclosing genes 36 and have a greater fitness cost 37 ., We next examined the distribution of intronic TEs in the sense orientation versus those in antisense with respect to the enclosing genes ( see Figure 3A for human and Figure 4A in Text S1 for mouse ) ., Since DNA transposons only comprise about 3% of both the human and the mouse genomes and almost all of them are ancient elements without evidence of any transposition activity during the past 50 Myr ( million years ) 1 , 2 , we excluded them from the following analyses to avoid uncertainties introduced by their relatively small numbers ., While previous studies have found an overall antisense orientation bias in genes ( particularly for LTR elements and LINEs ) 21 , 22 , we show here the existence of a much stronger bias in antisense for both LINEs and LTR elements near exons ., The excess of antisense TEs compared with sense elements near intron boundaries is probably the result of purifying selection , like the genome-wide orientation bias of TEs in genes ., This indicates in general that sense-oriented TEs near splice sites have a higher probability to influence normal gene transcription and are potentially more harmful to the host gene ., Interestingly , for SINEs we observed the same strong antisense bias in the mouse ( Figure 4A in Text S1 ) , but in the human genome we observed a sense orientation bias instead of antisense for SINEs at a close distance of 20–200 bp from exons ( Figure 3A ) ., These data are consistent with the Alu SINE study of Lev-Maor et al . 29 , in which the authors also observed more sense-oriented Alu elements near intron termini ., Since Alus account for two-thirds of human SINE elements and many antisense Alus possess a strong cryptic SA signal 13 , selection against antisense-oriented elements may explain the unusual underrepresentation of antisense oriented SINEs near splice sites in humans ., Furthermore , we also looked for evidence of any distributional bias of intronic TEs in terms of their proximity to either splice donor sites ( SDs ) or splice acceptor sites ( SAs ) ., We found the total numbers of elements near SA sites are much lower than SD sites for all three retrotransposon classes examined ( see Figure 3B for human and Figure 4B in Text S1 for mouse ) ., Since the core intronic splice signals at SD sites usually only consist of about 6 bp of terminal intron sequence compared with 20–50 bp at SA sites 30 , selection against physical disruption of critical splice motifs likely underlies this TE underrepresentation near SA sites ., Theoretically , harmful antisense transcripts of protein-coding exons may be generated by read-through transcription of antisense TEs near SD sites ., If such antisense transcripts have significant detrimental effects , then one might expect a larger proportion of TEs near SD sites to be in sense rather than in antisense due to purifying selection ., However , as shown in Figure 4A ( human ) and Figure 5A in Text S1 ( mouse ) , such predicted bias of sense orientated TEs near SD sites was not found except for human SINEs , which is likely explained by the fact mentioned previously that antisense Alus possess cryptic SA signals ., In fact , for other TE classes we observed more SD-associated elements oriented in antisense , probably indicating that antisense transcription is effectively silenced or not a general problem , and that sense oriented TE insertions are more detrimental ., The same analysis of TEs near SA sites revealed similar orientation bias patterns as for TEs near SD sites ., If the reduced frequency of TEs near intron boundaries reflects the force of selection against harmful insertions , one would predict that a higher fraction of mutagenic TEs in gene introns would be located within these TE underrepresentation zones ., To evaluate this prediction , we compiled information on documented intronic mutagenic TE insertions and examined their integration sites in introns ., Based on the TE activity and data availability , we focused on the following three TE families in our analyses: human Alu ( SINE ) , human L1 ( LINE ) and mouse LTR elements ., First , as the most abundant TE family , Alus have successfully propagated in the human genome and reached a total number of over one million copies 1 ., Even today , some of these elements are still active , generating new insertions and causing mutations linked to diseases 8 , 38 , 39 ., Based on the information provided by the dbRIP database ( http://dbrip . brocku . ca/ ) 27 , we found six de novo Alu insertions associated with human diseases within introns , all of which belong to the AluY subfamily ( the youngest subfamily of Alu ) and cause splice defects of the enclosing gene ( Table 1 in Text S2 ) ., Second , de novo disease-causing insertions of L1 , the active LINE family in humans , have also been reported 5 , 40 , 41 , 42 ., These elements play important roles in human retrotransposon-mediated pathogenesis because not only do they encode reverse-transcriptase ( RT ) and other proteins required for their own retrotransposition , but also for mobilizing Alus 43 ., In this study , our search of the dbRIP database identified a total of five intronic L1s associated with human diseases ( Table 2 in Text S2 ) , all of which cause transcriptional disruptions ., Last , since no mutagenic LTR insertions and only a few insertionally polymorphic ERVs or LTRs have been reported in human 4 , 6 , 44 , we turned to the mouse genome , where ERVs/LTR elements cause ∼10% of germline mutations , many of which have been well studied 7 ., In total we collected 40 cases of mutagenic LTR elements in mice: 15 from the Intracisternal A Particle ( IAP ) family , 18 from the Early Transposon/Mouse Type D retrovirus ( ETn/MusD ) family , and seven from other LTR elements or ERVs ., Again , all these ERV-induced intronic mutations in mice are due to transcriptional disruptions on the enclosing gene ( Table 3 in Text S2 ) ., For the three TE families listed above , we compared the intronic distribution of mutagenic elements with all full-length counterparts in the reference genomes and found highly consistent results ( Figure 5 and Table 2 ) ., As shown in Figure 5A , all six mutagenic Alu insertions are within the U-zone of SINEs ( i . e . <100 bp from the nearest exon ) , and all are oriented antisense with respect to the enclosing gene ., Moreover , five out of the six cases are near SA sites ., In comparison , only 1 . 83% of all full-length AluYs in the reference human genome are located within the 100 bp U-zone - strikingly lower than the mutagenic elements and also more than two-fold lower than that expected by chance ( p<2 . 2e-16; one-sample proportion test ) ., For all full-length AluYs within the U-zone we observed 47 . 7% elements in antisense , slightly lower than the random level ( 50% ) but much lower than mutagenic insertions ., Since intronic TEs show their strongest splice site bias when they are in extreme close proximity to an exon ( Figure 3B ) , we examined full-length intronic AluYs located less than 20 bp from exons and observed only 10% of such elements near SA sites ., Although we cannot directly compare this result to the case of mutagenic Alus due to their insufficient number within 20 bp to exons , the fact that five out of six mutagenic Alus are near SAs is noteworthy ., Similarly , Figure 5B shows that all five mutagenic L1 elements are within the U-zone for full-length LINEs ( i . e . <2 kb from the nearest exon ) ., Among them , four are sense-oriented and four are near SA sites ., In contrast , only 23 . 0% of full-length intronic L1s in the reference genome are within the U-zone , which is significantly lower than both the mutagenic L1s and our random simulation ( p<0 . 0004 and p<2 . 2e-16 , respectively; two-/one-sample proportion test ) ., Of those elements within the U-zone , only 27 . 7% are in sense , again significantly lower than both mutagenic insertions and the simulation ( p<0 . 035 and p<2 . 2e-16 , respectively; two-/one-sample proportion test ) ., Although the number of full-length L1s in the reference genome within 20 bp to exons is very limited , among a total of seven cases only two were found near SA sites ., We also examined the same parameters for mouse LTR elements ( Figure 5C and Table 2 ) ., As we expected , a high fraction of these mutagenic insertions ( 72 . 5% ) are within the U-zone of full-length mouse LTR elements ( i . e . <2 kb from the nearest exon ) ., More remarkably , all 15 mutagenic insertions from the IAP family were within the 2 kb U-zone ., Since the orientation information of some mutagenic LTR elements within the U-zone was not indicated in their original reports , we checked the remaining 26 cases and found 20 ( 76 . 9% ) were oriented in sense ., Among these mutagenic insertions in mice , five are located within 20 bp of exons , with three of them near SA sites ( 60% ) ., However , the situation is completely different for all full-length LTR elements in the sequenced mouse genome ( strain C57BL/6J , or B6 ) ., In contrast to mutagenic insertions , only 14 . 3% of full-length LTR elements in the reference genome were located within the 2 kb U-zone ( p<2 . 2e-16; two-sample proportion test ) , and of these elements only 30 . 1% are in the sense orientation ( p<2 . 65e-09; two-sample proportion test ) ., At a distance less than 20 bp to exons , we found six full-length LTR elements in the B6 reference genome but only one of them is near the SA site ( 16 . 7% ) ., In summary , the above analyses of mutagenic versus all full-length elements for the three retrotransposon families consistently showed an overrepresentation of mutagenic TEs within their respective U-zones but an underrepresentation of all full-length elements within the same regions ., Moreover , apparent differences in orientation and splice-site biases were also observed between mutagenic TEs and all full-length elements in the reference genomes ., These observations strongly suggest that intronic TE insertions within the U-zone have a much higher potential to be deleterious to the enclosing gene , particularly when oriented in antisense for human SINEs and in sense for LINEs and LTR elements ., When intronic TE insertions are in extreme proximity ( e . g . <20 bp ) to an SA site , they are very likely to be harmful and may cause functional abnormality of the enclosing gene ., We next extended our analyses to polymorphic AluY and L1 insertions not associated with any disease based on the dbRIP data ., These elements are considered as relatively young since they are not fixed in humans ., If , indeed , selection is still working upon these TEs , one might see an intermediate distribution pattern between that of mutagenic and all elements ., However , for both polymorphic AluYs and L1s we observed no significant differences from all full-length elements in the reference human genome ( data not shown ) ., While the limited total number of polymorphic insertions documented in dbRIP may partially account for this result , it is very likely that the distribution of these polymorphic TEs has already been shaped by selection ., However , for the youngest insertionally polymorphic mouse LTR elements , we have previously shown that they do have a distinct prevalence in introns and orientation bias compared with older elements 45 ., This suggests that some of these insertions are detrimental but have not been eliminated due to the artificial breeding environment of inbred strains 2 , 7 , 45 ., Indeed , some known detrimental LTR insertions have even become fixed in one or a few mouse strains 46 , 47 ., We therefore analyzed a list of polymorphic LTR insertions in four mouse strains from our previous study 45 , in which we had detected different distributions between polymorphic and common LTR elements ., Here we used polymorphic IAP and ETn/MusD elements that are present in only one of the four analyzed mouse strains ( presumed to be the youngest elements ) and found that 34 . 8% of intronic insertions were within the 2 kb U-zone ( Figure 5C and Table 2 ) , a fraction very close to the simulated prediction of a random distribution but significantly higher than all full-length LTR elements in the mouse reference genome ( 14 . 3%; p<5 . 58e-13; two-sample proportion test ) and lower than the mutagenic insertions ( 72 . 5%; p<9 . 79e-05; two-sample proportion test ) ., Moreover , we observed 23 . 2% of polymorphic LTRs in the U-zone as sense-oriented , which shows no statistical difference from that of all LTRs but is highly significantly lower than the mutagenic cases ( p<6 . 26e-07; two-sample proportion test ) ., Since our list of polymorphic LTR insertions in mice does not contain any intronic insertions within 20 bp of an exon , we could not perform the analysis of splice site proximity bias ., Nonetheless , the above observation of an intermediate distribution pattern of polymorphic insertions between mutagenic and all full-length TEs in the reference genome demonstrates that , indeed , purifying selection is the most likely underlying force shaping the observed intronic TE distribution patterns , and evidence suggests that such a process is ongoing ., If TEs within their respective U-zones are more likely to be harmful by causing splicing abnormalities , one can make two predictions ., One prediction is that TEs located in the U-zones would be associated with chimeric TE-gene transcripts more often than TEs located elsewhere in introns ., To test this prediction , we downloaded and analyzed the human expressed sequence tag ( EST ) data from the UCSC Genome Browser ( http://genome . ucsc . edu ) , in which only spliced transcripts were included ., We then screened for all spliced ESTs overlapping with intronic TEs ( i . e . chimeric ESTs ) ., As shown in Figure 6A , we observed that 11 . 7% of human SINE elements within the 100 bp U-zone are associated with chimeric ESTs ., In contrast , this ratio is only 1 . 6% for SINE elements outside the U-zone ., Similarly , for human LINEs in their 2 kb U-zone , we found 4 . 6% of them associated with chimeric ESTs , while outside the U-zone the ratio significantly drops to 0 . 7% ., Lastly , we identified 2 . 9% of human LTR elements as chimeric-EST-related in the 5 kb human LTR U-zone , but for elements outside the U-zone we observed only 0 . 9% ., All the above results are highly statistically significant ( all p-values<2 . 2e-16; two-sample proportion test ) , which reinforces the notion that TEs within their U-zones are more likely to be involved in aberrant splicing ., It should be pointed out , however , that the splicing events detected by this analysis are of unknown relevance and , indeed , because these TEs are fixed , are unlikely to have significant detrimental effects ., A second prediction is that TEs which were not eliminated from the U-zone would have weaker splicing signals compared with other TEs ., To examine this issue , we computationally analyzed potential splice sites within randomly selected solitary LTR sequences in human introns using NNSplice 48 ( see Materials and Methods ) ., As shown in Figure 6B , as the distance between the intronic LTR and its nearest exon decreases , the average number and the strength of predicted splice sites in these LTR sequences also decrease ., This observation indicates that LTRs carry fewer and weaker cryptic splice sites within the U-zone , especially when they are located in close proximity to exons ., While the above EST analysis suggests the importance of U-zones in TE-gene interactions , it would be useful to predict which particular intronic TEs are most likely to influence gene transcription based on their size , distance to the nearest exon , orientation , and proximity to particular splice site ., To conduct an initial evaluation of this concept , we examined a panel of polymorphic LTR element insertions in inbred mouse strains because they are currently highly active and , as discussed above , their genomic distribution suggests that some are likely detrimental but are maintained due to the artificial breeding environment ., In order to take the advantage of the available EST/mRNA data in the B6 reference genome , we restricted our set of intronic polymorphic LTR elements to those present in the B6 mouse strain 45 ., After excluding solitary LTRs and complex cases due to multi-gene families , we identified 44 full-length polymorphic LTR elements within the 2 kb U-zone ( data not shown ) ., We then inspected each region using the UCSC Genome Browser ( mouse genome version: mm9 ) to look for chimeric ESTs/mRNAs involving the LTR element and the enclosing gene and found such transcripts for 19 of the 44 genes ., For most of these 19 genes , the aberrant forms appear to be minor in abundance and it is difficult to estimate their overall impact on gene expression ., However , among these 19 genes , transcription of three of them ( Cdk5rap1 , Adamts13 , and Wiz ) has been shown to be significantly affected by the embedded LTR element 46 , 49 , 50 ., Judging from the frequency of annotated chimeric transcripts , two other genes among the group of 19 , Kcnh6 ( potassium voltage-gated channel , subfamily H ( eag-related ) , member 6 ) and Trpc6 ( transient receptor potential cation channel , subfamily C , member 6 ) , are of special interest ., While no evidence of transcriptional disruption caused by LTR element insertions has been reported in the literature for these genes , UCSC Genome Browser snapshots of their deposited mRNAs suggest significant involvement in the transcription of each gene ., For Trpc6 , two of seven mRNAs in the database terminate within a polymorphic IAP LTR element ( Figure 7A ) , and for Kcnh6 , one of three annotated mRNAs terminates within another IAP insertion ( Figure 7B ) ., Trpc6 plays an important role in vascular and pulmonary smooth muscle cells and its deficiency impairs certain allergic immune responses and smooth muscle contraction 51 ., Kcnh6 , also termed Erg2 ( eag related protein 2 ) , encodes a pore forming ( alpha ) subunit of potassium channels , and may serve a role in neural activation 52 ., To examine the potential effect of the IAP polymorphisms on transcription of these two genes , we first confirmed the presence or absence of these insertions by genomic PCR in a panel of mouse strains including B6 , A/J , and 129SvEv ., Indeed , an IAP is present in B6 and A/J but not in 129SvEv for the Trpc6 gene , and the IAP in the Kcnh6 gene is present only in B6 but not in A/J and 129SvEv ( data not shown ) ., Since both genes are highly expressed in the brain , we conducted quantitative RT-PCR on brain cDNA from all three mouse strains by setting one primer pair upstream of the insertion site and another primer pair flanking the insertion site , as indicated in Figure 7 ., In mouse strains carrying the IAP insertion , we found a significant decrease in the amount of normally spliced transcripts involving exons flanking the ERV insertion , compared with exons upstream of the insertion ., In contrast , we saw less difference between the upstream and flanking primer sets in strain ( s ) without the IAP insertion ( Figure 8 ) ., The blockage of normal Kcnh6 transcription is particularly striking , with very little normal splicing occurring for exons flanking the IAP in the B6 strain ., These data suggest significant transcriptional interference of these two genes mediated by the embedded IAPs , and it would be interesting to determine if this interference results in phenotypic differences between mouse strains with and without these insertions ., Over a million TEs have become fixed in human or mouse gene introns during evolution , and the vast majority of them presumably have no functional impact on the gene ., Yet , new disease-causing TE insertions do occur in introns and exert detrimental effects mainly by disrupting normal gene transcript processing ., The emergence of high throughput technologies has facilitated the discovery of an increasing number of TE germline polymorphisms and somatic insertions in human cancers , with the recent advances on studies of human L1 polymorphisms as the best example 23 , 24 , 25 , 26 ., However , little attempt has been made thus far to identify which of these polymorphic or somatically-acquired TEs may contribute to allele-specific gene expression differences and potential phenotypic variation or disease ., Methods are therefore needed to evaluate which TEs are most likely to affect gene transcription ., Here we have identified intronic underrepresentation zones near exons , where fixed TEs occur less often than expected by chance ., Strikingly , all documented human intronic Alu and L1 insertions and most mouse intronic LTR elements known to cause disease are located within these U-zones , strongly suggesting that TE elements in these locations are more likely to cause transcriptional disruptions and be eliminated by selection ., Moreover , TEs within their U-zones are more likely to be involved in spliced chimeric transcripts than those located elsewhere in introns , suggesting that some may be slightly detrimental ., Presumably in most of these cases the transcriptional effects must be insufficient to cause such insertions to be eliminated by purifying selection ., However , it is possible that even apparently subtle effects on gene splicing could have functional consequences ., On the other hand , previous studies have also demonstrated that TEs fixed in the host genome can participate in gene transcr

Introduction, Results/Discussion, Materials and Methods

Comprising nearly half of the human and mouse genomes , transposable elements ( TEs ) are found within most genes ., Although the vast majority of TEs in introns are fixed in the species and presumably exert no significant effects on the enclosing gene , some markedly perturb transcription and result in disease or a mutated phenotype ., Factors determining the likelihood that an intronic TE will affect transcription are not clear ., In this study , we examined intronic TE distributions in both human and mouse and found several factors that likely contribute to whether a particular TE can influence gene transcription ., Specifically , we observed that TEs near exons are greatly underrepresented compared to random distributions , but the size of these “underrepresentation zones” differs between TE classes ., Compared to elsewhere in introns , TEs within these zones are shorter on average and show stronger orientation biases ., Moreover , TEs in extremely close proximity ( <20 bp ) to exons show a strong bias to be near splice-donor sites ., Interestingly , disease-causing intronic TE insertions show the opposite distributional trends , and by examining expressed sequence tag ( EST ) databases , we found that the proportion of TEs contributing to chimeric TE-gene transcripts is significantly higher within their underrepresentation zones ., In addition , an analysis of predicted splice sites within human long terminal repeat ( LTR ) elements showed a significantly lower total number and weaker strength for intronic LTRs near exons ., Based on these factors , we selectively examined a list of polymorphic mouse LTR elements in introns and showed clear evidence of transcriptional disruption by LTR element insertions in the Trpc6 and Kcnh6 genes ., Taken together , these studies lend insight into the potential selective forces that have shaped intronic TE distributions and enable identification of TEs most likely to exert transcriptional effects on genes .

Sequences derived from transposable elements ( TEs ) are major constituents of mammalian genomes and are found within introns of most genes ., While nearly all TEs within introns appear harmless , some de novo intronic TE insertions do disrupt gene transcription and splicing and cause disease ., It is unclear why some intronic TEs perturb gene transcription whereas most do not ., Here , we examined intronic TE distributions in both human and mouse genes to gain insight into which TEs may be more likely to affect transcription ., We found evidence that TEs near exons are likely subject to strong negative selection but the size of the region under selection or “underrepresentation zone” differs for different TE classes ., Strikingly , all reported human disease-causing intronic TE insertions fall within these underrepresentation zones , and the proportion of TEs contributing to chimeric TE-gene transcripts is significantly higher when TEs are located in these zones ., We also examined insertionally polymorphic mouse TEs located within underrepresentation zones and found evidence of transcriptional disruption in two genes ., Given the growing appreciation for ongoing activity of TEs in human , our results should be of value in prioritizing insertionally polymorphic TEs for study of their potential contributions to gene expression differences and phenotypic variability .

genomics, genome evolution, evolutionary biology, gene regulation, molecular genetics, biology, computational biology, genomic evolution

journal.pgen.1000726

The Schizosaccharomyces pombe JmjC-Protein, Msc1, Prevents H2A.Z Localization in Centromeric and Subtelomeric Chromatin Domains

Chromatin is based on a repetitive structural unit called the nucleosome ., However the regulatory properties of chromatin are mediated by the differences between nucleosomes , due to post-translational modifications or presence of histone variants ., Cytologically , chromatin was initially divided into heterochromatin and euchromatin 1 ., The underlying molecular basis of this division was established at the nucleosomal level after the discovery of the partitioning of histone lysine methylations into hetero- and euchromatic domains 2 , 3 , 4 ., Further degrees of chromatin specificity have been revealed by studies of histone modifications and variants ., For example , trimethylation of histone 3 at lysine 4 ( H3K4me3 ) characterizes nucleosomes around RNAP II promoters 5 while incorporation of the histone 3 variant CENP-A characterizes nucleosomes of the inner centromere 6 ., How these differences arise and propagate , often at individual nucleosomes , is not clear , although clues are available ., For example , self-reinforcing feed-forward mechanisms can explain the propagation of nucleosomal states 7 , 8 , 9 ., These mechanisms rely upon a physical connection between a protein that binds a histone modification with an enzyme that catalyzes the same modification ., Notable examples include the association between H3K9 methyltransferase Clr4 and H3K9 methylation 10 , and Spp1 and Set1 for H3K4 methylation 11 ., Another way to maintain nucleosomal differences and chromatin domains are boundary mechanisms ., By blocking the spread of a self-reinforcing mechanism , boundaries such as those provided by insulators 12 or TFIIIC binding sites 13 restrict chromatin states to their respective domains ., Boundaries based on DNA cis elements are pre-fixed ., Other boundaries can be variably positioned depending upon expression levels of position effect variegation proteins , which enhance or diminish the spread of heterochromatin 14 ., However most explanations of self-reinforcing mechanisms and boundary phenomena assume that chromatin is one-dimensional ., Because it is obviously three-dimensional and apparently confined within a single cellular compartment , what mechanisms prevent the chaotic distribution of nucleosomal identities ?, This question is especially relevant for the processes that exchange canonical histones for histone variants ., After DNA replication , both daughter DNA molecules must be packaged in the same chromatin status as the parental molecule ., Canonical histone deposition occurs in a replication-coupled ( RC ) manner ., However , the deposition of certain histone variants occurs in a DNA replication-independent ( RI ) manner 15–18 ., For example , the H3 variant H3 . 3 is targeted to chromatin via an RI transcription-coupled mechanism 19 , 20 involving the H3 . 3-specific chaperone HIRA , as opposed to the RC chaperone CAF1 , which incorporates H3 . 1 21 ., RI chaperones are particularly susceptible to mistargeting of histone variants ., For example , the H3 variant CENP-A is enriched in the centromeric domain under guidance from neighbouring heterochromatin and epigenetic mechanisms 22 , 23 ., The histone chaperone RbAp48 interacts with CENP-A and is required for CENP-A loading ., However , RbAp48 interacts with both the CAF1 and HIRA chaperone complexes and can load either CENP-A or canonical H3 into chromatin in vitro 24 , 25 ., Furthermore , overexpression of CENP-A in various organisms leads to aberrant deposition in euchromatin 26 , 27 , and defects in CAF1 or HIRA nucleosome assembly pathways also lead to mistargeting of budding yeast CENP-A ( Cse4 ) 28 ., In this paper we focus on the H2A variant , H2A . Z , which is incorporated into budding yeast chromatin by Swr1 , the catalytic subunit of the Swr1 complex ( Swr1C ) and one of the SWI2/SNF2 superfamily of ATPase chromatin remodelers 29–33 ., Swr1C deposits H2A . Z-H2B dimers in chromatin both in vitro and in vivo , but does not remove H2A . Z from chromatin 32 ., In budding yeast , H2A . Z is mainly positioned at the promoters of lowly expressed or inducible genes 34–38 and is lost upon gene induction 30 , 31 ., At least some of these promoters show reduced induction in the absence of H2A . Z , suggesting that the destabilization of promoter nucleosomes by the inclusion of H2A . Z facilitates transcriptional initiation 35 ., H2A . Z may also be involved in defining chromatin boundaries and domains ., The absence of H2A . Z , or NuA4-mediated H2A . Z acetylation , allows telomeric gene silencing to spread beyond its usual domain resulting in the repression of sub-telomeric gene expression 39 , 40 ., H2A . Z has also been implicated in centromere function and chromosome segregation in mammals 41 , budding yeast 29 , 42 and fission yeast 43 , 44 , evident in increased rates of chromosomal loss in H2A . Z mutants and genetic interactions between H2A . Z and microtubule components 45 ., H2A . Z localizes to centric and pericentric chromatin in mammals 46 but was not found at centromeres in budding yeast 35 ., Hence H2A . Z and Swr1C are involved in many aspects of chromatin regulation ., Central to these processes is the incorporation of H2A . Z into specific nucleosomes ., However the basis for this specificity is unclear ., Here we report that this process is due to both positive and negative target selectivity by Swr1C , due in part to the JmjC-domain protein , Msc1 , which is a stoichiometric subunit of the fission yeast Swr1C ., Msc1 negatively regulates H2A . Z incorporation into specific chromatin regions at the inner centromere and sub-telomere ., As part of a study to develop datasets for comparative proteomics , we purified a S . pombe complex with high subunit orthology to the S . cerevisiae Swr1 complex 47 ., To characterize this complex in greater detail , we applied a sequential tagging strategy 48 to purify Swr1C via its Yaf9 , Swc4 , Swc2 and Msc1 subunits , as well as via Pht1 ( which is the fission yeast histone variant H2A . Z ) ., Notably , each of the tagged proteins , with the exception of Pht1 , appeared to be stoichiometric Swr1C subunits with no indication that any of them exist as free protein in the cell or as part of another complex ( Figure 1B and data not shown ) ., Msc1 is a JmjC domain protein , which has no orthologue in the S . cerevisiae Swr1C ., Msc1 is a member of the highly conserved Lid/Jarid1 family and has five zinc fingers , including one JmjN and three PHD fingers , an ARID/BRIGHT AT rich DNA binding domain and a Plu domain ( Figure S1 ) ., Msc1 was initially identified as a multi-copy suppressor of the absence of the cell cycle progression kinase , Chk1 49 , and has been recently linked with H2A . Z action 43 ., To investigate the role of Msc1 in Swr1C complex integrity , immunoprecipitations were performed using H2A . Z-TAP in an msc1Δ strain ., All Swr1C subunits except Msc1 were detected ., Therefore Msc1 is not required for the association of any other subunit or the association of Swr1C with H2A . Z ( Figure 1A and 1B ) ., Swr1 itself is essential for complex integrity , demonstrated by the absence of most Swr1C members in H2A . Z-TAP/swr1Δ and Msc1-TAP/swr1Δ purifications ., Notably the association of Swc2 and Swc5 in the H2A . Z-TAP/swr1Δ experiment indicates that these subunits directly bind H2A . Z ., Msc1 appears to be a stoichiometric subunit of Swr1C based on the intensity of its band in Coomassie stained PAGE gels , its presence in immunoprecipitations from multiple Swr1C baits and the ability of Msc1-TAP to pull down a complete Swr1C ., In addition to Swr1C , the H2A . Z-TAP purifications also yielded H2B , the Nap1/Nap1 . 2 histone chaperones , and the importin family protein Kap114 ( Figure 1 ) ., These proteins were not detected from Yaf9- , Swc4- , Swc2- , Msc1- or Swr1-TAP purifications but were detected in H2A . Z-TAP/swr1Δ , demonstrating they are H2A . Z-specific and do not interact directly with Swr1C but only with H2A . Z itself ., A similar interaction between H2A . Z and Nap1 in S . cerevisiae has been reported 30 , 32 ., To investigate the role ( s ) of Msc1 in H2A . Z metabolism , we performed genome-wide chromatin immunoprecipitation ( ChIP-chip ) analyses using myc-tagged H2A . Z in WT , msc1Δ and swr1Δ strains ., As shown for a representative euchromatic region ( Figure 2A ) , H2A . Z peaks were found predominantly at promoters in WT but were absent in swr1Δ strains ., In the absence of Msc1 , these peaks were found in the same places but often diminished ., To assess the genome-wide distribution of H2A . Z statistically , the tiling array data for every gene was represented by two values corresponding to the upstream intergenic region ( IGR ) and the open reading frame ( ORF ) ., At a cutoff of >1 . 5× , 660 IGRs showed enrichment for H2A . Z , indicating that about 1/7th of promoters in vegetative , exponentially growing , S . pombe contain strongly enriched H2A . Z ( Figure 2C ) ., This is a very similar value to S . cerevisiae 34 , 35 ., Furthermore only about 40% of all H2A . Z promoter peaks remained above the 1 . 5× threshold in the absence of Msc1 ( Figure 2C ) ., We further divided the occurrence of H2A . Z peaks into five categories with respect to mRNA expression level from very low to very high ( Figure 3A–3D ) ., In WT the H2A . Z peak corresponds with the first nucleosome in the transcribed region ( Figure 3A ) and the nucleosome-sparse promoter region can be seen as the low point in the H3 ChIP at −200 ( Figure 3C ) ., The H2A . Z peak does not correspond to the region of peak H3 density , which is found at +300 and presumably reflects peak nucleosomal density ., Furthermore , the most lowly expressed genes have higher H2A . Z peaks and the most highly expressed genes do not appear to have any H2A . Z at their promoters or elsewhere ., Loss of Swr1 abolishes the H2A . Z peak as expected ( Figure 3D ) , whereas loss of Msc1 results in a shift in all categories towards less H2A . Z , although the peak position remains the same ( Figure 3B ) ., Also notable is the absence of an H2A . Z peak at the −1 nucleosome , which is prominent in S . cerevisiae 5 but not Drosophila 50 ., We further evaluated the relationship between H2A . Z promoter peaks and gene expression levels to observe a strong negative correlation ( Figure 3E ) ., As expected from Figure 3A , H2A . Z occupancy inversely correlates with mRNA abundance ., However this inverse correlation does not apply to the least expressed genes ., Furthermore we observed strong positive correlations between H2A . Z peaks and H4K16 , H3K14 and other histone tail acetylations ( Figure S2 ) ., These proteomic and ChIP-chip data confirm that Swr1 and Swr1C are required for loading of H2A . Z into promoter sites in S . pombe euchromatin , whereas the role ( s ) for Msc1 are more subtle ., Msc1 is not required to specify the sites of H2A . Z loading , rather it contributes to H2A . Z occupancy either through loading efficiency or persistence ., Notably we also observed a strong inverse correlation between the genome wide distributions of H2A . Z and the nucleosome chaperones , Nap1 , Hrp1 and 3 51 , 52 ., Moving average plots show H2A . Z enrichment decreases with increasing Hrp1 , Hrp3 and Nap1 binding at IGRs across the genome ( Figure 3F ) ., IGRs bound by H2A . Z ( >1 . 5× cut-off ) do not coincide with IGRs bound by Nap1 and Hrp1 ( >1 . 8× cut-off ) ., In contrast , IGRs depleted in H2A . Z ( <0 . 60× ) show a strong overlap with IGRs bound by Hrp1 and Nap1 ( Figure 3G ) ., Considering that Nap1 physically interacts with H2A . Z , but not Swr1C , the absence of H2A . Z at Nap1- , Hrp1- and Hrp3-bound intergenic chromatin suggests that H2A . Z is removed from chromatin by Nap1 and the CHD remodelers ., Normally H2A . Z is absent from all centromeric regions , including both the CENP-A containing inner centromere and the pericentric heterochromatin ( Figure 4A ) ., However , in the absence of Msc1 or Swr1 , H2A . Z became incorporated specifically in the inner centromere ( Figure 4A , Figure S3 ) ., This corresponded to increased H3 ( Figure 4B ) and decreased CENP-A 43 occupancy ., Centromeric H2A . Z in the msc1Δ strain demonstrates that Msc1 acts as a negative regulator of H2A . Z inclusion or persistence at centromeres ., Additionally , the presence of centromeric H2A . Z in swr1Δ implies that H2A . Z does not strictly rely on Swr1C for loading into chromatin ., Similar to the centromeres , H2A . Z deposition at sub-telomeric domains was also affected by the losses of Msc1 and Swr1 ., In WT , H2A . Z is depleted from sub-telomeric domains ( approximately 100 kb in size ) at the left and right ends of chromosomes 1 and 2 ( Figure 5A , Figure S4 ) ., Loss of either Msc1 or Swr1 caused an increase of H2A . Z in these sub-telomeric domains ., The increase of H2A . Z was not as dramatic as that observed at centromeres and H2A . Z distribution did not adopt the euchromatic pattern of IGR promoter peaks , rather it was more scattered ., Notably the transition between euchromatin and the H2A . Z-free subtelomeric chromatin appears to be quite sharp on all four chromosome ends ( Figure 5A; Figure S4 ) ., The subtelomeric regions of chromosome 3 do not show H2A . Z depletion , or increased enrichment in the mutants , most likely because the rRNA gene repeats occupy both ends of this chromosome ( Figure S4 ) ., The sharp transition between euchromatin and sub-telomeric chromatin also corresponds to a transition of H3K4me2 levels 53 ., Notably this sharp transition coincides with the presence of LTR elements in at least two of the four cases ( Figure 5A , Figure S4 ) ., Genes residing in these sub-telomeric regions also tend to be the most lowly expressed 52 , 54–57 with an apparently sharp boundary corresponding to H3K4me2 and H2A . Z transitions ( Figure 5B ) ., Furthermore , at least for one subtelomeric region , Swi6 binding , which spreads from the densely H3K9 methylated telomeric region , appears to reach the same boundary 58 ., Based on these observations , we propose that the subtelomeric regions represent a distinct class of chromatin , and suggest the term ST-chromatin , which has different regional properties than bulk eu- or heterochromatin ., Examination of our genome-wide ChIP-chip datasets 52 further revealed that ST-chromatin is also depleted in H4K5 , H4K12 , H4K16 and H3K14 acetylation , and has a higher H3 density ., These regions are highly enriched for genes that are upregulated during meiosis , stress , and after the loss of Clr3 or Hrp1/Hrp3 ( Table 1 ) ., Like ST-chromatin , the inner centromeric ( IC ) domain is depleted in H3K4me2 compared to levels typically found in euchromatin 53 , ( see Figure 4D and Figure 5A ) ., Hence H3K4me2 and H2A . Z are similarly depleted at WT sub-telomeres and inner centromeres , and both chromatin domains display increased H2A . Z enrichment in swr1Δ and msc1Δ strains ., Gene expression changes in the absence of Msc1 , Swr1 and H2A . Z were measured by microarray analysis ., A significant overlap between the three datasets was found ( Figure 5C ) demonstrating that a common set of genes is affected in all three mutants ., In msc1Δ , very few genes were misregulated ( either up >1 . 5× , or down <0 . 67×; 85 ) compared to swr1Δ ( 265 ) or H2A . ZΔ ( 490 ) ., Genes up-regulated in all three mutant strains were lowly expressed in WT ., However loss of Msc1 had virtually no effect on the expression of any other genes , whereas loss of either H2A . Z or Swr1 also affected highly expressed genes ( Figure 5D ) ., The most striking observation from the expression profiling was increased expression in the mutant strains of many genes within approximately 160kb of the ends of chromosomes 1 and 2 ( Figure 5E ) ., H2A . ZΔ , msc1Δ and swr1Δ strains all showed significant up-regulation of sub-telomeric genes , despite having either complete loss ( H2A . ZΔ ) or increased sub-telomeric deposition ( msc1Δ and swr1Δ ) of H2A . Z ., The overlap between up-regulated genes in H2A . ZΔ , msc1Δ and swr1Δ was also higher at sub-telomeres than in the rest of the genome , indicating a similar loss of sub-telomeric transcriptional control in the three deletion strains ( data not shown ) ., In fact , more than 2/3rds of genes up-regulated in the absence of Msc1 lie in the sub-telomeric regions ., Notably , up-regulation spreads beyond the ST-chromatin boundaries , suggesting that the loss of ST-chromatin and its boundaries caused neighbouring effects ., As a further way to evaluate H2A . Z biology and Msc1 action , we developed quantitative mass spectrometry for fission yeast histone post-translational modifications including the histone variant , H2A . Z ., TAP-tagged H2A . Z was purified from WT , msc1Δ or swr1Δ strains with concomitant retrieval of associated H2B ( Figure S5 ) ., Unexpectedly , we found that the S . pombe H2A . Z N-terminal amino acid sequence was incorrect because the genome sequence was wrongly edited ( it has now been corrected ) ., The correct sequence is presented in Figure 6A with a comparison to other H2As ., H2A . Z has an extended N-terminal tail containing more lysines than canonical H2A ., Also , S . pombe H2A . Z includes two N-terminal methionines , which are either both present or absent , resulting in two variations of the N-terminal peptide ( named 1–22 or 3–22 respectively ) ., A comparison of absolute levels of peptides 1–22 and 3–22 revealed that 1–22 is the major isoform ., This isoform is always N-terminally acetylated ., About 2/3rds of total H2A . Z also carries 2 or more lysine acetylations ( Figure 6B and 6C , Figure S6 ) ., Hence the H2A . Z N-terminal tail is usually highly acetylated ., Multiple H2A . Z acetylation was reduced in msc1Δ and virtually abolished in swr1Δ strains ( Figure 6C and 6D ) ., Similarly , total H2A . Z levels were reduced by about 1/3rd in msc1Δ and about 4-fold in swr1Δ strains ( Figure 6E ) ., We combined Figure 6D and 6E to estimate the abundance of acetylated forms in WT , msc1Δ and swr1Δ strains ( Figure 6F ) ., Notably , the absolute amount of H2A . Z that was acetylated only on the N-terminus increased in both mutant strains , whereas all species of lysine acetylations were decreased ., In particular , lysine-acetylated H2A . Z almost vanished in the absence of Swr1 , whereas the level of N-terminal-only acetylated H2A . Z increased ., This near complete absence of multiply-acetylated H2A . Z coincides with the near complete absence of H2A . Z loading into chromatin in the absence of Swr1 ., Similarly , in the absence of Msc1 the reduction of multiply acetylated H2A . Z coincides with reduced H2A . Z occupancy , being approximately half in both cases ., This suggests that multiple acetylation of H2A . Z requires incorporation into nucleosomes and that there is a pool of unincorporated nuclear H2A . Z which is not multiply acetylated ., It also suggests that H2A . Z incorporated into nucleosomes in the absence of Msc1 is normally acetylated ., Like in budding yeast , H2A . Z incorporation into euchromatin in S . pombe depends on Swr1C and tends to be found at promoters of lowly expressed genes ., Apart from the most lowly expressed genes in vegetative growth , which are disproportionately found in subtelomeric regions 52 , 55–58 , there is a strong negative correlation between H2A . Z occupancy and mRNA expression level ., There is also a strong negative correlation between H2A . Z and Nap1/Hrp1/Hrp3 CHD remodeler occupancy 51 ., Because Nap1 binds to H2A . Z , we suggest that H2A . Z is loaded into many promoters and is removed by the CHD remodeler when the gene is expressed ., Hence we suggest that the observed H2A . Z distribution in a ChIP experiment is like a ‘snap-shot’ of expression levels and only partially reflective of the sites into which H2A . Z was loaded ., We propose that H2A . Z is loaded by Swr1C into the +1 nucleosome at most promoters and is subsequently removed by the Nap1/CHD remodeler upon transcription ., This suggestion concords with similar suggestions for budding yeast 34 , 35 and recent measurements of nucleosomal turnover , which occurs more rapidly at promoters 59 ., In euchromatin , loss of Msc1 had a quantitative but not qualitative effect on H2A . Z promoter occupancy ., It therefore appears that Msc1 does not play a role in defining the sites of H2A . Z deposition in euchromatin rather may contribute to the efficiency of reloading after Nap1/CHD removal in a transcription cycle ., By quantitative mass spectrometry , we found that H2A . Z is always N-terminally acetylated but variably acetylated on four lysines in the N-terminal tail ., In the absence of Swr1 , very little H2A . Z was found in chromatin and very little became multiply acetylated ., Furthermore , the N-terminally acetylated form of H2A . Z persisted regardless of the absence of Swr1 but overall H2A . Z levels were reduced , which equated with the absence of the multiply acetylated forms ., In agreement with similar suggestions from work with S . cerevisiae 39 , 60 , 61 , we conclude that lysine acetylation of H2A . Z depends upon loading into chromatin ., Notably , H2B associated with H2A . Z was heavily acetylated regardless of whether it was loaded into chromatin or not ( Figure S7 ) ., Consequently the two H2A . Z chaperones , Swr1C and Nap1/CHD may distinguish between free or loaded H2A . Z based on its acetylation status ( Figure 7 ) ., Msc1 is the largest of the seven JmjC domain proteins in fission yeast and we found it exclusively in Swr1C with no evidence that it occurs in any other complex or as free protein ., JmjC domain proteins have raised considerable interest recently because of their ability to demethylate lysines in histone tails 62 , 63 ., However a thorough bioinformatic analysis of JmjC domains indicated that Msc1 is probably not a demethylase because it lacks key residues in the catalytic domain 64 ., Msc1 is a member of the highly conserved Lid/Jarid1 family , which is based on a highly conserved architecture of seven protein domains arrayed in the same N- to C-terminal order ( Figure S1 ) ., This architecture indicates an integration of several conserved functions in addition to action by the JmjC domain ., In addition to Msc1 , S . pombe has another Lid/Jarid1 member , Lid2 , which was found in a complex with subunits of the Set1 H3K4 methyltransferase complex 65 and serves to regulate heterochromatin 66 ., Despite much recent activity , it remains unclear how JmjC proteins function to control chromatin ., Our proteomic data confines Msc1 function to H2A . Z ., Consequently Msc1 presents a good opportunity to understand the action of a JmjC protein ., The finding that the loss of Msc1 leads to ectopic incorporation of H2A . Z into the inner centromeric and subtelomeric chromatin was completely unexpected ., None of the known mechanisms for chromatin establishment or maintenance offer an explanation 67 ., These mechanisms are all based on cis-acting propagation of chromatin status , which directs the incorporation of new histones whether by RC or RI mechanisms 15–18 ., To our knowledge , the finding that Msc1 is required to exclude H2A . Z occupancy from two distinct chromatin domains is the first example of a mechanism that appears to prevent the incorporation of a histone variant into the wrong nucleosomes ., How Msc1 serves this role remains to be determined but it is notable that neither chromatin domain exists in budding yeast , which also does not contain an Msc1-like subunit in Swr1C ., Because H2A . Z incorporation into centromeric or ST chromatin does not require Swr1C , the simplest explanation involves Msc1 directing Swr1C to remove H2A . Z from these domains ., However other more complicated explanations are possible ., Because Msc1 has been described to be an E3 ubiquitin ligase 68 , possibly ubiquitinylation of H2A . Z plays a role in preventing incorporation or facilitating removal from these ectopic sites ., Recent work on another S . pombe JmjC/PHD finger protein , Epe1 , has identified roles in the maintenance of heterochromatin 69–71 , although the mechanism remains elusive ., It has been suggested that Epe1 is not a demethylase but a hydroxylase ( like the original JmjC/cupin domain protein , FIH; 70 , 72 ) ., This suggestion was supported by a consideration of conserved and non-conserved amino acids ., We note that Msc1 similarly lacks the important signature amino acids for demethylase activity but may retain some characteristics of hydoxylase activity ., Msc1 contains three different but highly conserved PHD fingers 73 ., PHD fingers encompass diverse functions 74 but many bind methylated or unmethylated lysines in histone tails 75–77 ., Hence many PHD fingers serve as ‘readers’ of the post-translational status of nucleosomes ., Similarly the JmjC domain , whether active or inactive as a lysine demethylase , also has the potential to read and possibly edit the post-translational status of lysine methylation in nucleosomes ., Hence Msc1 is well suited to regulate chromatin status in trans , especially to regulate the RI Swr1C histone chaperone ., We therefore suggest that other JmjC proteins , particularly the Lid/Jarid1 family , also serve to ‘read’ chromatin status and thereby convey information to regulatory processes ., H2A . Z is absent from sub-telomeric regions ( ca 80kb ) ., The transition from the normal euchromatic H2A . Z pattern to the sub-telomeric region appears to be sharp and coincides with an altered profile of H3K4me2 , the presence of retroviral insertions and also presumably the furthest limit of Swi6 binding and H3K9 methylation , which spread from the telomeres 58 ., We suggest the term ST-chromatin for this subtelomeric region to distinguish it from the densely H3K9 methylated heterochromatic telomeres and the H3K4me2 euchromatin of the chromosomal arms ., In addition to the lack of H2A . Z and uniformly lower levels of H3K4me2 , we also note that ST-chromatin is characterized by several distinct features including lower levels of H4K5/K12 acetylation than euchromatin and a higher density of H3 ( Table 1 ) ., Inner centromeric ( IC ) chromatin also has uniformly lower levels of H3K4me2 than euchromatin ., Hence it is possible that similarities between ST- and IC-chromatin , such as low H3K4me2 , account for the similar faulty incorporation of H2A . Z in the absence of Msc1 ., Notably , forced selection for neocentromere formation , after Cre recombinase centromere deletion , occurred in ST-chromatin 78 , and the authors favoured the explanation that the adjacent telomeric heterochromatin influenced the selection of the neocentromeric site ., In contrast , we suggest that the similarity between ST- and IC-chromatin is the primary reason ., This could be tested by Cre mediated deletion of the centromere on chromosome 3 , which has subtelomeric ribosomal repeats rather than ST-chromatin ( Figure S4 ) ., Because many meiotic specific genes are found in this domain , it appears that ST-chromatin is an example of regulation of a gene expression program by chromatin domain status ., Msc1 is required to maintain this status ., In its absence , many genes are derepressed ., Notably this derepression extends beyond the ST/euchromatin boundary into euchromatin ., Gene derepression in ST-chromatin was not only found in the absence of Msc1 or Swr1 , which provoke ectopic H2A . Z deposition into ST-chromatin , but also paradoxically in the absence of H2A . Z , which is normally absent from this domain ., This indicates that the maintenance of ST-chromatin requires euchromatic H2A . Z ., Furthermore gene repression in ST-chromatin requires Clr3 and Hrp1/3 ( Table 1 ) ., This evidence provides further reasons to conclude that the genes in ST-chromatin are coordinately regulated by chromatin status ., Swr1C purifications from TAP-tagged baits and mass spectrometry identification of complex members were carried out as described elsewhere 47 ., H2A . Z purification for MS was carried out according to the standard TAP-tag IP protocol , except bound material was eluted from IgG beads in 0 . 5M Na-acetate ( pH 3 . 4 ) and lyophilized to dryness ., Samples were reconstituted in HPLC buffer A ( 5% ACN + 0 . 1% TFA ) and separated by C4 RP-HPLC over a linear acetonitrile gradient ., Fractions were collected , lyophilized and digested with Arg-C protease for MS analysis ., Histones were purified using a protocol adapted from budding yeast 79 ., Briefly , harvested yeast pellets from 2L log-phase cultures were homogenized using a beadbeater ( BioSpec ) in a modified Nuclear Isolation Buffer ( 0 . 25M Sucrose , 60mM KCl , 15mM NaCl , 5mM MgCl2 , 1mM CaCl2 , 20mM HEPES pH8 . 0 , 0 . 5mM spermine , 2 . 5mM spermidine , 0 . 8% Triton X-100 , 10mM Na-butyrate and protease inhibitors ) ., The homogenate was centrifuged at 32 , 000g for 15 minutes , the crude chromatin pellet resuspended in 0 . 25N HCl , sonicated and rotated at 4°C for one hour ., Acid insoluble material was cleared by centrifugation and discarded ., Acid soluble material was purified in batch using BioRex70 ion exchange resin ( Biorad ) ., Samples were dialyzed against HPLC buffer A and separated by two rounds of RP-HPLC ( C4 and C18 ) over multi-step acetonitrile gradients ., Histone-containing fractions from the C4 separation were collected , re-separated over a C18 column , collected again , lyophilized , and digested with Arg-C for MS analysis ., Arg-C digested samples were first treated with propionic anhydride or deuterated acetic anhydride 80 and directly separated by C18 nanoLC according to standard conditions , and analysed on-line by an LTQ-Orbitrap mass spectrometer ( ThermoFinnigan ) ., Survey scans were conducted using the Orbitrap mass analyzer and MS/MS spectra acquired on the linear trap using a standard data dependent acquisition method ., Raw data was converted and submitted to MASCOT database searching including lysine methylation , dimethylation , tri-methylation , propionylation and acetylation as variable modifications ., Relative quantification of histone peptides was carried out using Xcalibur software ( ThermoFinnigan ) by extracting the areas of chromatographic peaks of the differentially modified parent ions ., ChIP was carried out as previously described 52 , 81 ., Immunoprecipitated DNA was amplified and hybridized to Affymetrix tiling arrays ., Microarrays were carried out in duplicate for both ChIPs and WT input ( Affymetrix GeneChip S . pombe 1 . 0FR Arrays ) at Pearson correlation coefficients of r>0 . 97 ., Probes are tiled for both strands of the genome at an average of 20 base pair resolution ., Antibodies used were against H3 ( ab1791 , Abcam ) and H2A . Z-myc ( 9E10 , ab10826 , Abcam ) ., Expression arrays were carried out as described 82 ., Raw data from Affymetrix ( . CEL format ) was analyzed by Affymetrix Tiling Analysis Software ( TAS ) v1 . 1 using quantile normalization plus scaling and assigned with a bandwidth of 100 ., The data was normalized with DNA input and each probe was assigned to the S . pombe genome ( September edition 2004 , Sanger center UK ) coordinates in TAS ., Visualization of data was performed using the Affymetrix Integrated Genome Browser ( IGB ) ., The resulting linear ratio was extracted for each probe position , defined as the center ( 13th ) base coordinate for each 25-nucleotide probe ., Data sets from ChIP on chip experiments were used to map all coding genes onto an average gene , using a similar method as previously described 83 , 84 ., Briefly , we used the upstream intergenic region and part of coding region for each gene ., The analyzed region was −800 to +2800 bp from respectively the start codon of the gene with a 20bp resolution ., Values for each probe were attributed to the closest assigned position ., Gene expression data was normalized to genomic DNA fragmented by DNAse1 57 using TAS and the genes were assigned in

Introduction, Results, Discussion, Materials and Methods

Eukaryotic genomes are repetitively packaged into chromatin by nucleosomes , however they are regulated by the differences between nucleosomes , which establish various chromatin states ., Local chromatin cues direct the inheritance and propagation of chromatin status via self-reinforcing epigenetic mechanisms ., Replication-independent histone exchange could potentially perturb chromatin status if histone exchange chaperones , such as Swr1C , loaded histone variants into wrong sites ., Here we show that in Schizosaccharomyces pombe , like Saccharomyces cerevisiae , Swr1C is required for loading H2A . Z into specific sites , including the promoters of lowly expressed genes ., However S . pombe Swr1C has an extra subunit , Msc1 , which is a JumonjiC-domain protein of the Lid/Jarid1 family ., Deletion of Msc1 did not disrupt the S . pombe Swr1C or its ability to bind and load H2A . Z into euchromatin , however H2A . Z was ectopically found in the inner centromere and in subtelomeric chromatin ., Normally this subtelomeric region not only lacks H2A . Z but also shows uniformly lower levels of H3K4me2 , H4K5 , and K12 acetylation than euchromatin and disproportionately contains the most lowly expressed genes during vegetative growth , including many meiotic-specific genes ., Genes within and adjacent to subtelomeric chromatin become overexpressed in the absence of either Msc1 , Swr1 , or paradoxically H2A . Z itself ., We also show that H2A . Z is N-terminally acetylated before , and lysine acetylated after , loading into chromatin and that it physically associates with the Nap1 histone chaperone ., However , we find a negative correlation between the genomic distributions of H2A . Z and Nap1/Hrp1/Hrp3 , suggesting that the Nap1 chaperones remove H2A . Z from chromatin ., These data describe H2A . Z action in S . pombe and identify a new mode of chromatin surveillance and maintenance based on negative regulation of histone variant misincorporation .

Chromatin is based on a repetitive structural unit called the nucleosome ., However , the regulatory properties of chromatin are mediated by the differences between nucleosomes , due to post-translational modifications or the inclusion of histone variants ., These differences are maintained by inheritance through cis-acting epigenetic mechanisms ., Here we describe a case where the local character of chromatin is not only determined by cis-acting mechanisms but also negatively regulated in trans ., The case involves loading of the histone H2A variant , H2A . Z , into chromatin ., We show that H2A . Z in the yeast Schizosaccharomyces pombe is mainly found in genes at the first transcribed nucleosome and is inserted into this nucleosome by the Swr1C remodeling machine ., However , Swr1C has a regulatory subunit , Msc1 , which is not required for H2A . Z promoter loading but prevents H2A . Z occupancy in the inner centromere and subtelomeric regions ., These two specialized regions are neither eu- nor heterochromatin and share certain characteristics , which may predispose them to the aberrant inclusion of H2A . Z and the requirement for trans regulation by Msc1 .

molecular biology/histone modification, genetics and genomics/gene expression, molecular biology/centromeres, genetics and genomics/epigenetics, chemical biology/protein chemistry and proteomics, molecular biology/chromatin structure

journal.pgen.1006995

Environmental perturbations lead to extensive directional shifts in RNA processing

Variation in gene expression has long been associated with cellular and organismal phenotypes ., For example , studies have found that gene expression in blood and bronchial epithelial cells differs among individuals with asthma 1 , 2 , 3 , 4 ., Such differences in gene expression occur in specific cellular pathways , such as the glucocorticoid response pathway 1 , 5 , 6 , 7 , leading to the general usage of glucocorticoids to treat asthma ., These studies , and others , have demonstrated that variation in gene expression plays a role in complex traits and cellular responses 8 , 9 , 10 , 11 , 12 ., More recently , however , researchers have begun to assess the impact of alternative mRNA isoform usage on phenotypes ., Previous studies have found that RNA processing , leading to differential isoform usage , is different in certain diseases such as Alzheimer’s disease and several forms of cancer 13 , 14 , 15 , 16 , 17 ., Furthermore , studies have identified global shifts in exon usage associated with developmental or diseased cellular states ., For instance , shorter 3’ untranslated region ( UTR ) isoforms are prevalent in proliferating or cancerous cells 18 , 19 ., Cancer is also associated with increased retention of introns 20 , 21 ., Li et al . recently identified genetic variants associated with inter-individual variation in mRNA splicing and identified almost 2 , 900 splicing Quantitative Trait Loci ( QTLs ) ., Further , they showed that splicing QTLs are also enriched for genetic variants associated with several complex traits in Genome-Wide Association Studies ( GWAS ) , demonstrating the potential importance of splicing misregulation in complex traits 22 ., Previous work from our lab and others have shown that gene-by-environment interactions can impact both gene expression and complex traits 23 , 24 , 25 , 26 , 27 , 28 ., While splicing QTLs have been identified both in humans and mice 22 , 29 , 30 , 31 , less is known about how gene-by-environment interactions may affect RNA processing ., The first step to address this question is to characterize RNA processing in response to environmental perturbations ., RNA processing is regulated in response to certain environmental stimuli , such as cancer therapy drugs , nutrient starvation and infection 32 , 33 , 34 , 35 some of which influence cell viability 36 , 37 , 38 ., For example , UV exposure leads to differential isoform usage in the gene BCL2L1 , which is involved in the regulation of apoptosis ., UV leads to increased abundance of Bcl-xs which favors apoptosis as opposed to Bcl-xl which is anti-apoptotic 39 ., Other studies have demonstrated widespread , directed changes in the regulation of RNA processing ., Infection with Listeria monocytogenes and Salmonella typhimurium led to increased inclusion of cassette exons and shorter 3’UTRs genome-wide 35 ., The longer versions of 3’UTRs that were shortened were found to be enriched with particular microRNA binding sites , suggesting that the RNA processing shift leading to shorter 3’UTRs may be a way for these genes to evade down-regulation following infection ., Despite the fact that these studies have increased our understanding of factors that influence changes in RNA processing , they have investigated only a limited number of environments ., Cataloguing and characterizing RNA processing changes across many environments , in a tightly controlled study using specific treatments , is necessary to increase our understanding of the cellular mechanisms leading to variation in RNA-processing , including which aspects are common across many environments and which are specific to certain perturbations ., Our study aimed to systematically assess the impact of a broad range of environmental perturbations on the regulation of RNA processing ., We measured RNA processing patterns in five cell types across over 30 treatments , corresponding to a total of 89 cellular environments with 3 biological replicates and additional technical replicates ( 297 RNA-seq libraries in total with 130M reads per library on average ) 23 ., The treatments represent compounds to which we are exposed in daily life , ranging from metal ions and vitamins to allergy medication ., This work catalogs the extent of alternative RNA processing in response to a wide range of specific environmental perturbations and provides evidence for molecular mechanisms by which trans factors influence this process ., Using high-throughput RNA sequencing , we identified 32 compounds that induce gene expression changes in 32 , 451 genes in 5 different cell types ( a total of 89 environments ) 23 ( S1 Table ) ., In order to identify changes in RNA processing , we utilized the probabilistic framework implemented in the software Mixture of Isoforms ( MISO ) 40 , which characterizes changes in exon usage by calculating a percent spliced in ( PSI , Ψ ) value ., The Ψ value is calculated by taking the ratio of reads specific to an inclusion isoform—specifically , reads aligning to the alternative exon or its junctions—to all reads that can be mapped to the region ( including constitutive exons ) ., Instead of entire isoforms , which may involve multiple RNA processing mechanisms that are convolved together , we focused on individual exons that are tied to known RNA processing mechanisms ., We focused on events that involve known curated isoforms ( see Methods ) , rather than novel isoforms , and characterized variation in RNA processing events across different environments ., This allowed us to learn about cis- and trans-acting mechanisms leading to the RNA processing response ., Specifically , we characterized changes in eight event types: skipped exons ( SE ) , retained introns ( RI ) , alternative 3’ or 5’ splice sites ( A3SS , A5SS ) , mutually exclusive exons ( MXE ) , alternative first or last exons ( AFE , ALE ) , and tandem untranslated regions ( TandemUTR ) ( Fig 1A , S1 Fig shows a treatment color key used throughout the manuscript ) ., Across all conditions , we identified 15 , 300 changes in RNA processing , representing a unique set of 8 , 489 events that significantly differ between at least one treatment and control condition ( Table 1 , S2 Table ) ., These events are found in genes enriched for gene ontology terms such as RNA binding , gene expression , metabolic process , response to stress and cell cycle suggesting their role throughout the cellular response to environmental perturbation ( BH FDR < 5% , S3 Table ) 41 ., Each significant change in an RNA processing event was identified based on RNA sequencing data across cell lines derived from three unrelated individuals ( example in Fig 2 and at http://genome . grid . wayne . edu/RNAprocessing ) ., Across all environments , the most abundant event types with shifts were RI , AFE and ALE ( relative to the number of sites tested ) , while the least abundant was A3SS ( Fig 1B ) ., As we studied events in a given cell type across conditions , we found treatment-specific shifts in RNA processing , resulting in vast differences in the number and type of event shifts ( examples in Fig 3 ) ., We found a wide range in the number of significant shifts across treatments , with vitamin D producing the highest number ( 2 , 530 events ) and BP3 leading to the lowest number of significant shifts in RNA processing ( 65 events ) ( average = 478 events , 0 . 5% of events tested ) ( Fig 1C ) ., The number of RNA processing changes in each environment is minimally correlated to the sequencing depth of the library ( Spearman’s ρ = 0 . 15 , p = 0 . 02; S2A Fig ) , but it is correlated to the number of differentially expressed genes in each environment ( Spearman’s ρ = 0 . 59 , p = 1 . 22 × 10−8; S2B Fig ) , suggesting the same underlying mechanism inducing changes in RNA processing and in overall gene expression ., In addition to differences in the overall number of RNA processing changes , we also found differences in relative number of changes in certain event types ( Fig 1B ) ., While changes in AFEs represent the greatest overall number of changes across environments , there is substantial variation in the extent to which each event type changes within each treatment ( Fig 1B ) ., We utilized a generalized linear model to determine the proportion of event types among significant event shifts in a given treatment ., With this model , we identified 3 treatments that showed enrichment for an event type , including vitamin E , tunicamycin , and cadmium ., For example , vitamin E is enriched for A5SS while cadmium is depleted for RI ( Fig 1B ) ., Together , these results demonstrate that , similar to changes in gene expression , a large number of RNA processing events change in response to environmental perturbations ., Regulation of RNA processing events in response to environmental perturbations may be mediated by trans factors that impact many RNA processing events of the same type , or by cis-acting regulatory sequences that would impact each event separately ., To investigate these two mechanisms we considered global shifts in RNA processing ., Among the 8 event types with changes following treatment , 5 were considered directionally: SE , RI , AFE , ALE , and TandemUTR ., Specifically , for each event , ΔΨ was assigned a sign to indicate a qualitative difference between treatment and control conditions that is consistent across all events ., We used a positive ΔΨ ( same as positive Z-score ) to indicate either an increase in usage of the skipped exon , upstream AFE , downstream ALE , longer TandemUTR or intron retention in the treatment sample as compared to control ( Fig 4A ) ., This allowed us to consider transcriptome-wide trends across sites that may indicate a shift in overall regulation of RNA processing , such as consistent inclusion of an exon ., When we focused on treatments with at least 30 significant RNA processing shifts of a certain event type , 19% of treatments showed a correlation ( p < 0 . 05 ) between changes in RNA processing and changes in gene expression ( examples in S3 Fig , S4 Table ) ., For example , iron induced a positive correlation between ALE and gene expression ( Spearman’s ρ = 0 . 27 , p = 0 . 002 ) ( S3A Fig ) ., Specifically , genes shifting towards usage of the downstream ALE following iron treatment also have increased expression in the treatment samples ., On the other hand , selenium leads to the opposite effect: increased expression following selenium is found in genes that utilize the upstream ALE ( Spearman ρ = −0 . 18 , p = 1 × 10−4 ) ( S3B Fig ) ., These data suggest that in specific environments , cells respond with concerted shifts in RNA processing events and gene expression ., However , while we did identify correlations between RNA processing events and gene expression in some conditions , the absence of strong correlation in many of the conditions suggests that other factors play a role in RNA processing shifts ., To investigate this possibility , we started by examining similarities of shifts across sites that might suggest certain factors that play a role in the cellular response to environmental perturbation ., We investigated whether the global shifts in events had consistent direction across environments suggesting a shared trans-acting mechanism of change ., First , we found that 8 environments led to an enrichment for SE shifts toward either inclusion or exclusion of the alternative exon ( two-sided , binomial test compared to the expected proportion of 50% , p-value < 0 . 05 , Fig 4B ) ., Specifically , six environments were enriched for positive SE shifts which indicate global inclusion of the alternative exon while two led to more negative shifts or exclusion of the exon ., When studying RI across all environments , we identified 20 environments that lead to global shifts in intron inclusion ., Specifically , 18 out of 20 were enriched for positive events ( p-value < 0 . 05 , Fig 4B ) , thus showing enrichment for intron retention as compared to the control for most environments ., These results suggest a common mechanism for intron retention in cells that respond to changes in the environment ., For example , even though vitamin D causes many more changes in alternative splicing in PBMCs , all cell types trend towards retaining introns following vitamin D treatment ., This can be more clearly seen when considering all RI events ( not just significant events ) , where all 4 cell types show a shift toward more positive values in their ECDF ( Kolmogorov-Smirnov ( KS ) test p < 0 . 05 for 3 of 4 cell types ) denoting higher ΔΨ values , retaining of introns , following vitamin D treatment ( S4A and S4B Fig ) ., Of the 5 event types whose direction could be assessed , AFE global shifts are observed in the most environments ( Fig 4B ) ., Specifically , 34 environments led to usage of the downstream AFE ( negative Z-scores ) , while seven treatments were significantly enriched for shifts to the upstream AFE ( positive Z-scores ) ( p-value < 0 . 05 , Fig 4B ) ., Interestingly , several treatments lead to opposite AFE shifts in different cell types demonstrating the importance of the cellular background in response to environmental perturbations ., For example , insulin leads to a shift toward the downstream AFE in SMCs but a shift toward the upstream AFE in melanocytes ., This is also apparent when we consider all event shifts in these environments ( KS test p < 0 . 05 ) ( S4C and S4D Fig ) ., Both ALE and TandemUTR also showed deviation from the expected 50:50 ratio of positive to negative events but the trend was less clear ( S5 Fig ) ., These results demonstrate that while there are similar trends in the proportion of significant events across event types in a given environment , the directionality of the event shift is often different ., Furthermore , these results show that global shifts in RNA processing events can be determined solely by the treatment or by the combined effect of treatment and cell type ., In order to elucidate the specific factors involved in the global shifts in SE and RI events , we focused on factors likely to influence RNA processing , specifically splicing factors ., We quantified the gene expression changes of splicing factors across all environments to determine if there was a correlation to the number of positive ( inclusive ) RNA processing shifts ., The underlying hypothesis is that shifts in exon usage may be explained by splicing factors that:, 1 ) have activity largely mediated by changes in gene expression , and, 2 ) have the same influence over splicing in all treatments ., We identified 14 splicing factors ( of 166 tested ) with changes in gene expression correlated with percent significant positive events of all significant events for RI ( BH FDR < 5% , example in Fig 5A and 5B , S5 Table ) ; none were found for SE ., Notably , we identified that changes in expression of LARP7 are positively correlated with RI events ., This suggests that the increased expression of LARP7 under treatment conditions leads to more intron retention ( positive RI events ) ., Previous work has shown that LARP7 promotes skipping of alternative exons 42 ., Our results suggest that LARP7 also plays a role in intron retention events ., This trend can be seen across all environments considered ., For example , selenium leads to an increase in expression of LARP7 and more intron retention ., The lack of splicing factors correlated with SE could be due to several factors ., First , unlike RI , there are multiple treatments that do not have global trends towards either inclusion or exclusion of the skipped exon ., This may indicate that each exon is controlled by unique mechanisms and so searching for a particular responsible splicing factor may not be the best model ., Furthermore , across treatments , we do not see the same coordinated changes as we do for RI further hindering our ability to identify a splicing factor across treatments ., Many factors may influence RNA processing differently following various treatments and so we may miss an effect by investigating common expression patterns across environments ., Also , some factors are known to have different effects depending on binding location and not necessarily on overall gene expression ., For example , when SR proteins ( serine-arginine proteins ) bind upstream of 5’ splice site , they induce splicing but do not have the same effect when bound in the intron 43 ., With this in mind , we asked whether predicted binding sites of splicing factors may explain SE and RI ., First , we characterized motifs that are present upstream , downstream or within the alternative unit ( exon for SE or intron for RI ) ., We , then , utilized an elastic-net regularized generalized linear model ( GLM-NET ) to predict splicing changes in 5 environments with greater than 100 significant event shifts ( 3 with SE , 2 with RI ) , based on the binding motif occurrences of splicing factors ., When studying the model as a whole , we found that area under the curve ( AUC ) for each environment ranges from 0 . 67 for melanocytes exposed to loratadine to 0 . 87 for PBMCs exposed to vitamin D , suggesting that binding of splicing factors is important for determining changes in splicing following treatment , but the impact differs across cellular environments ( Fig 5C ) ., We also found that the genomic location of a binding site , relative to the splicing event , is an important predictive feature ., For example , a motif for RBM8A ( M054_0 . 6 from RNAcompete 44 ) is a part of the predictive model of SE in PBMCs treated with vitamin D but only when the motif is located in the upstream intron ., This demonstrates the positional effect of binding that others have characterized for some splicing factors 43 , 45 , 46 , 47 and expands its importance across a large number of environmental perturbations ., We hypothesized that transcription factors regulate AFE shifts and TSS usage in response to environmental perturbations ., Similar to our analysis with splicing factors , we first hypothesized that shifts in AFE could be the consequence of changes in gene expression for transcription factors that promote usage of either the upstream or the downstream TSS and have similar effects in all environments ., We identified 328 ( out of 1 , 342 ) transcription factors whose change in expression is correlated with shifts in AFE ( BH FDR < 5% ) ( example in Fig 5D and 5E , S6 Table ) ., Together , these results suggest that transcription factor binding influences the choice of TSS leading to a consequent shift in alternative first exon usage ., To directly determine the effect of transcription factor binding on AFE shifts , we then utilized transcription factor footprints identified in DNase-seq data from ENCODE and the RoadMap Epigenomics 49 , 48 , 50 to predict shifts in AFE usage in 14 environments ., We used footprints from more than 150 cell types to better capture a wider range of cellular environments , as determined by tissue of origin or culturing conditions ., To predict AFE shifts , we considered the number of footprints present within 1000bp in either direction of each transcription start site ( defined as the beginning of each alternative first exon ) , and used GLM-NET ( as we did in the splicing factor analysis ) ., Across the 14 environments , the AUC ranges from 0 . 71 for cadmium in LCLs to 0 . 92 for dexamethasone in LCLs ( Fig 5F ) ., These data suggest that transcription factor binding predicts changes in AFE following treatment ., By inducing changes in transcription factor binding , specifically by perturbing the cellular environment , we can validate the effect of binding on AFE usage ( Fig 6A ) ., To this end , we performed ATAC-seq in LCLs following treatment with selenium and its vehicle control ., First , we noticed that selenium leads to an overall reduction in chromatin accessibility near transcription start sites ( Fig 6B ) ., To determine how selenium influences binding of transcription factors near alternative TSS , we characterized chromatin accessibility following treatment with selenium , or control ( with the footprints used in the prediction analysis ) ., We found significant differences in chromatin accessibility for 64 motifs , near the TSS that was preferentially used in the treatment versus the TSS preferred in the control condition ., Of these 64 motifs , 26 are ETS transcription factor family members ( or from motifs with similar sequence preferences ) ., The most significant motif was for the transcription factor ELF2 ( also in the ETS family , p-value = 5 . 4 × 10-6 ) ., We found a global decrease in chromatin accessibility at the ELF2 motif but there was a milder decrease in accessibility at the preferred TSS following selenium ( Fig 6C and 6D ) , compared to the non-preferred TSS ., These data suggest that at baseline ELF2 promotes transcription at both TSSs ., However , following selenium treatment , though there is an overall decrease in ELF2 from both TSSs , there is a greater decrease from one TSS and this leads to a shift towards less usage of that TSS following treatment ., All 26 motifs predicted from the ETS family of transcription factors show a similar change in binding as ELF2 ., More broadly , these results support a mechanism for changes in TSS usage driven by changes in chromatin accessibility and potentially transcription factor binding in response to perturbations of the cellular environment ., To further validate the effect of ELF2 binding on AFE usage , we characterized AFE across 373 unrelated , European individuals from the GEUVADIS data 51 ., We identified 8 , 263 AFE events that can be characterized in at least 200 individuals ., Using these data we performed AFE quantitative trait loci ( QTL ) analysis , by focusing on the SNPs in ELF2 footprints in the cis region of an AFE ., We found an enrichment for QTLs in SNPs that were also predicted 48 to impact binding of ELF2 , compared to those that do not affect binding ( Fisher test p-value < 0 . 05 , OR = 3 . 14 , Fig 6E ) ., For example , the G allele of a SNP in IGHMBP2 ( rs546382 ) is predicted to promote binding of ELF2 and the genotype of this SNP is associated with Ψ values across the GEUVADIS dataset ( p-value = 1 . 1 × 10−35 , Fig 6F ) ., In this way , using genetic perturbation , we were able to validate the impact of transcription factor binding , and specifically binding of ELF2 , on AFE usage ., We describe 15 , 300 event shifts following a wide range of environmental perturbations at 8 , 489 unique RNA processing event sites ., We have provided a browsable web-resource cataloguing these RNA processing shifts ., Researchers interested in a given gene , isoform , or treatment will be able to access our data to determine when RNA processing shifts occur and which other genes respond under similar environments ., Mining of our results has the potential to inform on the mechanisms by which a cell responds to environmental perturbations and its genome-wide effect on RNA processing ., Interestingly , we identified some RNA processing changes that occurred across biologically-related treatments ., For example , of the 1 , 030 significant RNA processing shifts occurring in PBMCs , 120 can be found to shift in the same direction among various metal ion treatments ( copper , iron , molybdenum , zinc and cadmium ) ., Additionally , even though the COS inhibitors ibuprofen , aspirin and acetaminophen are structurally distinct and likely have different mechanisms of action , of the 147 event shifts in melanocytes , 25 shifts are shared across these treatments ., In addition to informing on particular treatments and cell types , these data can be utilized to study similarities of cellular responses across the wide range of treatments used here ., While these data are valuable for understanding cellular response , further research is necessary in order to verify these RNA processing shifts in vivo ., The majority of events could be characterized as AFE , ALE , RI or SE , suggesting that these are the most influenced by the environmental perturbations considered here ., This is distinct from previous reports that TandemUTR events change most following infection 35 and suggests diverse mechanisms through which the cells respond to their environment ., Previous work has studied the role of splicing factors and transcription factors in RNA processing , in the absence of specific environmental perturbations ., For example , others have shown that multiple splicing factors influence cassette exon usage , several of which fall into 2 protein families: hnRNPs and SRSFs ., These 2 protein families often result in opposite splicing patterns 43 ., These proteins may play a role in several of the RNA processing events that we study here , including SE , RI , A5SS , and A3SS ., There are other studies that characterized proteins related to polyadenlyation site usage ( which we study as TandemUTR ) , including E2F , CSTF2 , CSTF64 52 , 53 ., Furthermore , recent studies have suggested that binding of transcription factors may influence differential use of transcription start sites in mice 54 ., While these studies demonstrate the role of trans factors in RNA processing , we aimed to determine their role in global RNA processing changes in response to environmental perturbation ., Across 89 cellular environments , we found that binding sites for specific trans factors predict the shifts in events following treatment , thus demonstrating the importance of these factors and their binding locations for cellular response ., We often find that not all binding sites for a given motif are predictive , but rather only binding sites in a certain location relative to the exon of interest ., Furthermore , while previous studies have demonstrated the impact of binding location on RNA processing events at baseline , we demonstrate that the effect of binding in a certain location is treatment-specific ., These results highlight the importance of studying trans factor binding across various environments ., Further analysis of these binding sites will aid in understanding the details of the molecular mechanisms regulating RNA processing response to each cellular environment ., For example , motifs associated with weaker binding of a trans factor may allow for more rapid changes in RNA processing and a more rapid cellular response ., Previous reports have characterized differences in transcription factor expression and binding across cellular environments 49 , 55 ., Here , we show that variation in transcription factor binding following environmental perturbations may determine TSS usage in addition to their function of influencing total gene expression ., Feng et al . demonstrated the influence of transcription factors on TSS usage in mice 54 ., Here , we expand on this knowledge by showing a similar function in human cells , both in response to many environmental changes and across individuals ., Using ATAC-seq data , we further pinpointed factors such as ELF2 whose binding is disrupted by the environment , leading to changes in TSS usage ., The changes in TSS usage are also observed when binding is disrupted by genetic variation in the GEUVADIS data , as shown in the AFE QTL analysis ., Transcription factors are often regulated by environmental changes and are then responsible for impacting expression of many genes to promote re-establishment of cellular homeostasis ( reviewed in 56 ) ., Therefore , we suggest that TSS usage may also play a substantial role in cellular response and homeostasis ., Alternative RNA processing is predicted to occur in over 95% of multi-exon genes in humans across various tissues 57 ., Our comprehensive catalog of genome-wide RNA processing changes can be utilized in future studies that aim to understand the role of RNA processing under various conditions and diseases as many of the treatments we used represent compounds to which individuals are commonly exposed ., Furthermore , because RNA processing is associated with complex trait variation 22 , 17 , individual differences in RNA processing , specifically in response to environmental changes , could shed light on variation in organismal phenotypes ., We used deep-sequenced RNA-seq data ( fastq files ) from Moyerbrailean et al . , 2016 23 ., Briefly , five cell types ( LCL , PBMC , HUVEC , melanocyte and smooth muscle cells ) were treated with 50 compounds to which humans are regularly exposed ., Each environment ( cell type and treatment ) was represented in cell lines derived from three , unrelated individuals ., We utilized the step 2 sequencing data which focused on 89 environments , with at least 80 differentially expressed genes , ( S1 Table ) that were sequenced with 150bp reads to an average of 130M reads/library ( 297 RNA-sequencing libraries ) ., These 89 environments include treatments and three vehicle controls ( S1 Table ) ., In order to detect alternative splicing , we used Mixture of Isoforms ( MISO ) 40 , which requires reads of the same length ., Therefore , we selected reads with a length greater than or equal to 120bp ., All reads were trimmed to 120bp ., We also removed reads whose paired end was less than 120bp ., Reads were aligned to the hg19 human reference genome using STAR 58 ( https://github . com/alexdobin/STAR/releases , version STAR_2 . 4 . 0h1 ) , and the Ensemble reference transcriptome ( version 75 ) with the following options:, STAR --runThreadN 12 --genomeDir <genome>, --readFilesIn <fastqs . gz> --readFilesCommand zcat, --outFileNamePrefix <stem> --outSAMtype BAM Unsorted, --genomeLoad LoadAndKeep, where <genome> represents the location of the genome and index files , <fastqs . gz> represents that sample’s fastq files , and <stem> represents the filename stem of that sample ., Each of the 297 RNA-sequencing libraries were sequenced multiple times in Step 2 23 in order to obtain adequate coverage ., These sequencing runs were merged using samtools ( version 2 . 25 . 0 ) ., We further removed reads with a quality score of < 10 ( equating to reads mapped to multiple locations ) ., In order to detect alternative splicing , we used MISO on samples aligned as above ., Each control compound ( ethanol , water or DMSO ) was used to treat the same individual cell line in three technical replicates ., The reads from each of these samples were combined to perform the following analysis ., We utilized the events annotated and listed on http://miso . readthedocs . io/en/fastmiso/index . html ., Specifically , we searched our data for 8 types of events with 5 from version 2 ( SE , RI , A5SS , A3SS , and MXE , http://miso . readthedocs . io/en/fastmiso/annotation . html ) and 3 from version 1 ( AFE , ALE and TandemUTR , 59 ) ., Two versions were used because AFE , ALE and TandemUTR were not annotated in version 2 ., We then ran miso . py on each of our samples for each of the 8 event types ., miso . py --run indexed_events/ my_sample1 . bam --output-dir my_output1/, --read-len 120, Then , we used summarize_miso . py to get the summary statistics for each event in each sample , including the percent spliced in value ( PSI , Ψ ) ., summarize_miso --summarize-samples my_output1/ summaries/, To identify differential splicing , we used compare_miso . py which compares each event between treatment and control samples in the same individual cell line and experimental batch ( plate ) ., compare_miso --compare-samples my_output1/ my_control1/ comparisons/, This script resulted in a ΔΨ , a Bayes factor and p-value for each comparison ., We then focused on comparisons where both treatment and control contained 2 reads covering each isoform uniquely and a total of 10 reads unique to either isoform for SE , RI , A5SS , A3SS , MXE , AFE and ALE ., TandemUTR can only have reads specific to one isoform as the other isoform is simply a shorter version and completely overlaps the first ., Therefore , we focused on comparisons of TandemUTR where both treatment and control contained 5 reads specific to the longer isoform and 10 total reads that covered either isoform ., Additionally , in order to inform on a cut-off

Introduction, Results, Discussion, Methods and materials

Environmental perturbations have large effects on both organismal and cellular traits , including gene expression , but the extent to which the environment affects RNA processing remains largely uncharacterized ., Recent studies have identified a large number of genetic variants associated with variation in RNA processing that also have an important role in complex traits; yet we do not know in which contexts the different underlying isoforms are used ., Here , we comprehensively characterized changes in RNA processing events across 89 environments in five human cell types and identified 15 , 300 event shifts ( FDR = 15% ) comprised of eight event types in over 4 , 000 genes ., Many of these changes occur consistently in the same direction across conditions , indicative of global regulation by trans factors ., Accordingly , we demonstrate that environmental modulation of splicing factor binding predicts shifts in intron retention , and that binding of transcription factors predicts shifts in alternative first exon ( AFE ) usage in response to specific treatments ., We validated the mechanism hypothesized for AFE in two independent datasets ., Using ATAC-seq , we found altered binding of 64 factors in response to selenium at sites of AFE shift , including ELF2 and other factors in the ETS family ., We also performed AFE QTL mapping in 373 individuals and found an enrichment for SNPs predicted to disrupt binding of the ELF2 factor ., Together , these results demonstrate that RNA processing is dramatically changed in response to environmental perturbations through specific mechanisms regulated by trans factors .

Changes in a cell’s environment and genetic variation have been shown to impact gene expression ., Here , we demonstrate that environmental perturbations also lead to extensive changes in alternative RNA processing across a large number of cellular environments that we investigated ., These changes often occur in a non-random manner ., For example , many treatments lead to increased intron retention and usage of the downstream first exon ., We also show that the changes to first exon usage are likely dependent on changes in transcription factor binding ., We provide support for this hypothesis by considering how first exon usage is affected by disruption of binding due to treatment with selenium ., We further validate the role of a specific factor by considering the effect of genetic variation in its binding sites on first exon usage ., These results help to shed light on the vast number of changes that occur in response to environmental stimuli and will likely aid in understanding the impact of compounds to which we are daily exposed .

gene regulation, regulatory proteins, dna-binding proteins, dna transcription, transcription factors, sequence motif analysis, research and analysis methods, sequence analysis, selenium, genome complexity, genomics, bioinformatics, proteins, gene expression, chemistry, rna splicing, biochemistry, rna, rna processing, nucleic acids, database and informatics methods, genetics, biology and life sciences, physical sciences, computational biology, introns, chemical elements

journal.pgen.1000660

A Genome-Wide Association Analysis Identified a Novel Susceptible Locus for Pathological Myopia at 11q24.1

Myopia is a refractive error ( http://en . wikipedia . org/wiki/Refractive_error ) of the eye in which parallel rays of light focus in a plane anterior to the retina resulting in blurred vision ., Myopia is one of the most common ocular disorders worldwide , and is in much higher prevalence in Asians than in Caucasians ., Recent population-based surveys in the elderly reported that the prevalence of myopia was approximately 25% in the Caucasian populations 1 and 40% in the East Asian ( Chinese and Japanese ) populations 2 , 3 ., Myopia is divided into two distinct subsets , namely , common and pathological myopia ., Pathological myopia , also called high myopia , is distinguished from common myopia , also called low/moderate myopia , by excessive increase in axial length of the eyeball , which is the most important contributor to the myopic refraction 4 , 5 ., The axial length of the eyeball in adults is approximately 24 mm , and its elongation by 1 mm without other compensatory changes results in a myopic shift of −2 . 5 to −3 . 0 diopters ( D ) ., It has been shown that distribution of the axial lengths of the adult myopic population is bimodal 6 , and the subgroup with elongated axial length in the bimodal distribution corresponds to pathological myopia ., This group comprises 1% to 5% of the population 3 , 7 , and is commonly defined by axial length greater than 26 . 0 mm which is equivalent to refractive errors greater than −6 D 8 ., The excessive elongation of the eyeball causes mechanical strain with subsequent degenerative changes of the retina , choroid , and sclera ., The degenerative changes at the posterior pole of the eye such as chorioretinal atrophy or posterior staphyloma are clinically important and unique to pathological myopia 9 ., These unique degenerative changes at the posterior pole result in uncorrectable visual impairment due to decreased central vision and make pathological myopia one of the leading causes of legal blindness in developed countries 10–13 ., It has been reported that not only environmental factors , such as near work and higher education , but also genetic factors contribute to the development of myopia , in particular , of pathological myopia 14 ., Previous twin studies reported that the estimated heritability of refractive error and axial length is up to 0 . 90 15 , 16 , although that might be overestimated due to common environmental effects 17 ., Multiple family-based whole genome linkage analyses of myopia reported at least 16 susceptible chromosomal loci ( MYP1–16 in OMIM database; 10 loci for pathological myopia 18–27 and 6 for common myopia 28–30 ) ., Among them , at least 8 chromosomal loci , such as 12q21–23 ( MYP3 ) , 22q12 ( MYP6 ) and 2q37 . 1 ( MYP12 ) were successfully validated by at least two independent studies 31 , 32 ., However , no genes responsible for the disease have been identified ., The genome-wide association ( GWA ) study using single nucleotide polymorphisms ( SNPs ) as markers is an alternative approach to identify genetic risk factors of common diseases ., This approach has been successfully applied to identify genetic risk factors for multigenetic diseases including ophthalmic diseases such as age-related macular degeneration 33 , 34 and exfoliation syndrome 35 ., To identify the genetic risk factors of pathological myopia , we conducted a two-stage GWA-based case/control association analysis using 411 , 777 markers with 830 Japanese patients and 1 , 911 Japanese controls ( 297 cases and 934 controls in the first stage , and 533 cases and 977 controls in the second stage ) ., A total of 839 pathological myopic patients with axial length greater than 26 . 0 mm in both eyes were enrolled in the current study ., In order to maximize the detection power , patients with axial length greater than 28 . 0 mm in both eyes were enrolled in the first stage of genome scan ., No other clinical features were accounted for the assignment of patients to either stage ., 824 out of 839 patients ( 98 . 2% ) had degenerative changes specific to pathological myopia ., Other features of cases and controls who passed quality control procedures of genotyping results ( see Materials and Methods ) were summarized in Table 1 ., For the first stage , we scanned the genome of 302 cases using the Illumina HumanHap550 BeadChip , which launches 561 , 466 relatively frequent SNPs ( minor allele frequency>0 . 05 ) distributed across the human genome at an average interval of 6 . 5 kilobases ( kb ) ., Five cases and 149 , 689 SNPs were excluded due to quality control criteria ( see details in Materials and Methods ) and genotyping results of 411 , 777 SNPs in autosomes for 297 cases were used for the statistical analysis ., They were compared with 934 controls from the JSNP database 36 for association with phenotype using χ2 test for trend ., Genomic Control ( GC ) method 37 revealed only a slight inflation of the test statistics ( GC parameter λ\u200a=\u200a1 . 068 ) ., We identified 29 SNPs in 22 chromosomal regions with P-value adjusted by GC being smaller than 10−4 ( Figure 1 and Table S1 ) ., Among them , seven SNPs at chromosome 8p12 were in strong linkage disequilibrium ( LD ) and likewise two SNPs at chromosome 10q22 . 2 ( pair-wise D′>0 . 95 and r2>0 . 9 ) ., Thus , we selected one representing SNP from each region and tested 22 SNPs in the second stage ., For the second stage analysis , 537 cases and 980 population controls were genotyped by Taqman method ., Among them , four cases and three controls were excluded due to low call rates ( <90% ) ., Genotyping success rates of the 22 SNP markers in the remaining 1 , 510 samples were greater than 96 . 8% ., The genotype counts of the first and second stages were combined for meta-analysis ., One SNP , rs577948 , showed a strongly suggestive association ( P\u200a=\u200a2 . 22×10−7 ) ( Table 2 ) in the meta-analysis whereas the remaining 21 SNPs were not significant ( P>10−5 ) ( Table S1 ) ., The SNP rs577948 which showed P\u200a=\u200a2 . 22×10−7 by meta-analysis with OR of 1 . 37 ( 95% confidence interval ( CI ) : 1 . 21–1 . 54 ) for the risk allele ( nominal P\u200a=\u200a2 . 80×10−5 and P\u200a=\u200a1 . 42×10−3 in the first and second stages , respectively ) ( Table 2 ) was located at chromosome 11q24 . 1 ( Figure 2A ) ., Using the results of the first stage , an LD block which extended a 55-kb region containing rs577948 was generated ., Six additional SNP markers within the block were included in the genome scan chip ( Figure 2B ) ., Among them , we selected three markers with adjusted P-value smaller than 0 . 01 in the first stage for further genotyping by Taqman method with DNAs used for the second stage ., Weaker associations than that of rs577948 were obtained for these three markers by meta-analysis ( Table 2 ) ., As shown in Figure 2B , two genes were located in a 200-kb region containing rs577948 ., BLID is a cell death inducer containing BH3-like motif 38 , which is located approximately 44-kb upstream of rs577948 ., The other gene , LOC399959 , is a hypothetical non-coding RNA 39 which encompassed 114-kb DNA in the region , and rs577948 is located in its second intron ., BLID is known as a cell-death inducer expressed in cytoplasm , in mitochondria at lower abundance , and in various human cancer cells from different tissues 38 ., LOC399959 was reported as a hypothetical non-coding RNA with a relatively ubiquitous expression pattern ., We assessed the expression of the genes by RT-PCR using cDNAs of human retina and brain and those of HeLa cells as positive control ., Expressions of both genes were detected in human retinal tissue as well as in human brain and HeLa cells ( Figure 3 ) ., Myopic refraction and axial length are reported to be a complex trait under polygenic control in which contribution of each gene is relatively small 40 ., In the current study , two-stage GWA analysis identified a region at chromosome 11q24 . 1 , in which rs577948 showed strongly suggestive P\u200a=\u200a2 . 22×10−7 with OR of 1 . 37 ( 95% CI: 1 . 21–1 . 54 ) for the allele G . Our GWA study identified only one strongly-suggestive locus ., This may principally be due to the sample size of our study not being adequate ., Recent genetic studies of complex traits with higher prevalence enroll much larger number of samples ., In contrast , recruitment of patients with pathological myopia is difficult due to its lower prevalence , particularly those with degenerative changes ( namely degenerative myopia ) ., In order to improve insufficient detection power , we assigned pathological myopia patients with longer axis ( greater than 28 . 0 mm ) to the first stage ., This strategy might be the reason we were successful in identifying the candidate region with relatively small number of cases ., Insufficiency of detection power due to a limitation in sample number may be a reason for difference between the findings of preceding linkage studies and ours ., OMIM database lists 10 MYP regions ( MYP1–5 , 11–13 , 15 and 16 ) for pathological myopia 18–27 and 6 MYP regions ( MYP6–10 and 14 ) for common myopia 28–30 ., None of these 16 MYPs are on chromosome 11q ., Stambolian and colleagues reported heterogeneity LOD score of 1 . 24 at 11q23 in their linkage study for common myopia in Ashkenazi Jewish descent , which is the closest locus to our region reported to date 29 ., Because the linkage signal was not strong and the band 11q23 ( chr11 , position 110 , 000 kb to 120 , 700 kb in the NCBI database ) is more than 800 kilobases apart from our LD block in 11q24 . 1 ( chr11 , position 121 , 535 kb to 121 , 590 kb ) , whether or not they overlap each other is inconclusive ., On the other hand , our study did not identify the associated SNPs in any of MYPs ., Although the insufficiency of detection power may be a reason for difference between our study and the linkage studies , there are other possible reasons ., In general , any difference in the study designs could cause heterogeneous results ., Firstly , there are two definitions of pathological myopia based on two distinct criteria , namely , the axial length and refractive error ., In the current study , we enrolled pathological myopic patients based on the axial length ( greater than 26 . 0 mm in both eyes ) , and not on the refractive error commonly used in the previous studies ( refractive errors greater than −6 D ) ., We focused on patients with vision-threatening degenerative changes 9 and the axial length fits better than refractive error for our purpose ., The mean refraction in our myopic patients was −13 . 14±4 . 57 D ( eyes that had undergone cataract surgery or corneal refractive surgery were excluded from this calculation ) which indeed correspond to pathological myopic group in the previous linkage studies ., On the other hand , it is not clear whether the patients enrolled in the linkage studies fulfill our criteria because the distribution of axial length and degenerative phenotypes in the cases are unknown ., The difference in definition of pathological myopia may result in different susceptibility loci between studies ., Secondly , the methodology used is different between studies , namely , linkage analysis and association analysis using linkage disequilibrium mapping ., The results of linkage and association studies of complex genetic traits are often different ., Family-based linkage analysis is much more suitable for identifying rare genetic variants with large effects whereas SNP-based GWA analysis is more powerful in detection of relatively common variants with smaller effects in complex diseases 41 ., Finally , the difference can also be due to the ethnicities of the samples enrolled ., In the current study , all cases and controls were Japanese ., Only one genome-wide linkage study has previously been published for pathological myopia in Japanese 42 and the others were for non-Japanese populations ., It would be interesting and important to examine the association of our locus in other ethnicities ., Ethnic variations in disease susceptibility genes have been reported in various genetic traits including ophthalmological disorders ., One such example is an SNP in the complement factor H gene ( rs1061170 ) which has a large effect size with age-related macular degeneration in Caucasians 33 , 43 , 44 but much smaller in East Asian populations due to a remarkably lower risk allele frequency ( ∼35% in Caucasians and ∼5% in East Asians ) 45 ., Another example is exfoliation syndrome and LOXL1 where the risk allele of rs1048661 is inverted between Icelandic ( allele G ) and Japanese ( allele T ) populations 35 , 46 ., Because of a large variation in prevalence of myopia among ethnic groups , a future trans-ethnic investigation of myopia risk genes will be important to dissect genetic backgrounds underlying the etiology of myopia ., Although the susceptibility locus contains BLID and LOC399959 , it seems premature to discuss the involvement of LOC399959 in myopia since it is a hypothetical non-coding gene ., BLID plays a proapoptotic role involving the BH3-like domain by inducing a caspase-dependent mitochondrial cell death pathway 38 ., Indeed , several animal and pathological studies suggested the functional role of apoptosis in pathological myopia 47 , 48 ., Moreover , a recent genome-wide linkage study followed by a fine-scale association mapping identified a myopia susceptibility gene locus containing the PARL gene which inhibits the mitochondrial pathway of apoptosis by interaction with OPA1 49 ., In this context , BLID seems functionally relevant with the pathogenesis of pathological myopia ., However , the true functional origin of association in this region has yet to be determined by further detailed investigation along with replication studies to validate our findings ., All procedures used in this study conformed to the tenets of the Declaration of Helsinki ., The Institutional Review Board and the Ethics Committee of each institution approved the protocols used ., All the participants were fully informed of the purpose and procedures , and a written consent was obtained from each ., Japanese pathological myopic cases were recruited at the Center for Macular Diseases of Kyoto University Hospital , the High Myopia Clinic of Tokyo Medical and Dental University , and Fukushima Medical University Hospital ., All subjects underwent comprehensive ophthalmologic examinations , including dilated indirect and contact lens slit-lamp biomicroscopy , automatic objective refraction evaluation , and measurement of the axial length by applanation A-scan ultrasonography ( UD-6000 , Tomey , Nagoya , Japan ) or partial coherence interferometry ( IOLMaster , Carl Zeiss Meditec , Dublin , CA ) ., As a general population control of the first stage , genotype count data of 934 healthy Japanese subjects were obtained from the JSNP database 36 ., For the second stage , 980 healthy Japanese individuals were recruited at Aichi Cancer Center Research Institute ., Genomic DNAs were extracted from peripheral blood leukocytes with QuickGene-610L DNA extraction kit ( FUJIFILM Co . , Tokyo , Japan ) ., We designed to scan the genome in two stages ., A total of 839 patients and 1 , 914 controls were separated into two groups; 302 cases and 934 controls for the first stage , and 537 cases and 980 controls for the second stage ., In order to increase the detection power , patients with longer axis of the eyeball ( greater than 28 . 0 mm ) were principally assigned to the first stage ., For the first stage analysis , 561 , 466 SNPs were genotyped in 302 patients of pathological myopia using Illumina HumanHap550 chips ( Illumina Inc . , San Diego , CA ) ., This chip covers approximately 87% of the common genetic variations in the Asian population 50 ., Cluster definition for each SNP was performed using Illumina BeadStudio Genotyping Module ., A systematic quality control procedure of the genome scan results was applied as follows ., Samples were evaluated for data quality first and markers were subsequently excluded ., Genetic proximity of sample pairs was evaluated with pi-hat in PLINK 51 and four samples with indication of kinship or sample duplication were excluded ., Genotypes in X chromosome were used for checking the precision of the phenotype record , and only one sample was removed due to mismatch in gender ., The final sample size of pathological myopia was 297 ., As a population-based control , genotype count data by the genome scanning of 934 healthy Japanese subjects using the same chip were obtained from the JSNP database 36 ., The chip contained 515 , 154 markers in autosomes that are common in the cases and controls ., We excluded 78 SNPs due to low successful call rate ( <95% ) in the cases , 1 , 760 SNPs due to the distortion of Hardy-Weinberg Equilibrium ( HWE ) in the controls ( P<10−3 by HWE exact test ) and 46 , 722 monomorphic SNPs ., 54 , 817 SNPs with minor allele frequency less than 0 . 05 in both cases and controls were also excluded ., After these quality control procedures , a total of 411 , 777 SNPs were used for the statistical analysis ., The genotyping call rate was greater than 97 . 43% ( median call rate 99 . 99% ) for DNA sample and 98 . 21% ( median call rate 100% ) for SNP marker ., Association between genotypic distribution of each SNP and the disease was examined using a χ2 test for trend ., The OR and the 95% CI were estimated using Woolfs method 52 ., Inflation in the test statistics was assessed using the genomic-control method 37 ., Haploview 53 software was used to infer the LD in the targeted regions ., SNPs with P-value adjusted by genomic control being smaller than 10−4 were selected as candidates for second stage ., Among the candidate SNPs , LD indices ( D′ and r2 ) were calculated with Haploview and when multiple SNPs were in strong LD ( D′>0 . 95 and r2>0 . 9 ) , one representative SNP was chosen to be genotyped in the second stage ., In the second stage , 537 cases and 980 controls were genotyped with the Taqman SNP assay using the ABI PRISM 7700 system ( Applied Biosystems , Foster City , CA ) ., The 302 pathological myopic cases in the first stage were also genotyped to validate the concordance between Illumina Infinium assay and Taqman assay ., Samples with low successful call rate ( <90% ) were excluded from the study ., Subsequently four cases and three controls were excluded and data of 533 cases and 977 controls were used for the analysis ., The concordance rate ranged between 98 . 68% and 100% for the 22 SNPs ., The genotype counts of the first and second stages were combined for meta-analysis using the Mantel-Haenzel method 54 as a fixed-effect model ., The OR heterogeneity between the first stage and the second stage was evaluated using Cochrans Q-statistic P-value ., The data from the second stage were also evaluated for association independently from the first stage ., Human retina cDNAs were obtained from Takara Bio Inc . ( Kyoto , Japan ) ., Total RNA of HeLa cells and human whole brain were also obtained from the same manufacturer and cDNAs were synthesized using the First-Strand cDNA Synthesis Kit ( GE Healthcare Life Sciences , Piscataway , NJ ) ., Two pairs of oligonucleotides were synthesized for RT-PCR; 5′-TTGGGTTCCAACAAAGAACC-3′ and 5′-CTTTTACAGGGCCTCAGCAG-3′ for BLID , and 5′-GGCGACATCAGACAGACAGA-3′ and 5′-AGGACCAGCTGAAAGGAACA-3′ for LOC399959 ., Expression of glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) was tested for cDNA quantification using 5′-GACAACAGCCTCAAGATCATCA-3′ and 5′-GGTCCACCACTGACACGTTG-3′ ., PCR reactions were performed under the following condition: initial denaturation at 96°C or 2 minutes , followed by 35 cycles ( for BLID and LOC399959 ) or 18 cycles ( for GAPDH ) at 96°C for 20 seconds , 60°C for 40 seconds , and polymerization at 72°C for 40 seconds .

Introduction, Results, Discussion, Materials and Methods

Myopia is one of the most common ocular disorders worldwide ., Pathological myopia , also called high myopia , comprises 1% to 5% of the general population and is one of the leading causes of legal blindness in developed countries ., To identify genetic determinants associated with pathological myopia in Japanese , we conducted a genome-wide association study , analyzing 411 , 777 SNPs with 830 cases and 1 , 911 general population controls in a two-stage design ( 297 cases and 934 controls in the first stage and 533 cases and 977 controls in the second stage ) ., We selected 22 SNPs that showed P-values smaller than 10−4 in the first stage and tested them for association in the second stage ., The meta-analysis combining the first and second stages identified an SNP , rs577948 , at chromosome 11q24 . 1 , which was associated with the disease ( P\u200a=\u200a2 . 22×10−7 and OR of 1 . 37 with 95% confidence interval: 1 . 21–1 . 54 ) ., Two genes , BLID and LOC399959 , were identified within a 200-kb DNA encompassing rs577948 ., RT–PCR analysis demonstrated that both genes were expressed in human retinal tissue ., Our results strongly suggest that the region at 11q24 . 1 is a novel susceptibility locus for pathological myopia in Japanese .

Myopia is one of the most common ocular disorders with elongation of axis of the eyeball ., Pathological myopia or high myopia , a subset of myopia which is characterized with excessive axial elongation and degenerative changes of the eye , is a leading cause of visual impairment ., Since genetic factors play significant roles in its development , identification of genetic determinants is an urgent and important issue ., Although family-based linkage analyses have isolated at least 16 susceptible chromosomal loci for pathological or common myopia , no gene responsible for the disease has been identified ., We conducted the first genome-wide case/control association study of pathological myopia in a two-stage design using 411 , 777 markers with 830 Japanese patients and 1 , 911 Japanese controls ., We identified a region strongly suggestive for the disease susceptibility at chromosome 11q24 . 1 containing BLID and LOC399959 ., Their expression was confirmed in human retina with RT–PCR ., BLID encodes an inducer of apoptotic cell death , and apoptosis is known to play an important functional role in pathological myopia ., We believe that our study contributes to further dissect the molecular events underlying the development and progression of pathological myopia .

ophthalmology, genetics and genomics/complex traits, ophthalmology/retinal disorders

journal.pcbi.1005071

Inference of Ancestral Recombination Graphs through Topological Data Analysis

Since the publication of the first draft of the human genome 1 , 2 , there has been an explosion in genomic data ., The genomes of thousands of different human individuals have been sequenced 3 , several hundreds of eukaryotic genomes have been characterized , and new viral , bacterial and archaeal species are being sequenced on an almost daily basis 4 , 5 ., Darwin provided a historical dimension to the taxonomical enterprise , proposing that closely related species in the hierarchical taxonomy share ancestors ., Since then , tree-like structures have been proposed to represent the evolutionary/historical relationship between organisms ., In the last few years , however , the richer and more comprehensive genomic characterization of many organisms have underscored the need of representations that are not strictly tree-like ., Phenomena such as horizontal gene transfer in bacteria 6 , the ability of viruses to borrow and lend genes across species , and hybridization in metazoa ( in plants , in particular 7 , 8 ) are exposing some of the limitations imposed by tree-like phylogenetic structures ., The definition of species itself becomes cumbersome in bacteria and viruses 9 ., Within many species , including humans , genetic recombination is so pervasive that tree-like representations are useless ., It is then natural to wonder what other frameworks could be used to capture phylogenetic relationships without losing the interpretability and simplicity of trees 10–12 ., Of particular interest are representations that reduce to trees when evolution is tree-like; that capture genetic relations between ancestors , and identify genomic regions originating from different ancestral lineages; and , more generally , that allow for an interpretation of the observed data in terms of a chronological sequence of events ., Several such frameworks have been proposed in the last two decades ., The study of phylogenetic networks has been an area particularly active 13–15 ., Phylogenetic networks provide representations that extend trees to graphs ( networks ) , generating loops when the data does not fit into a tree ., Some of those methods can easily be applied to more than one hundred genomes 16–21 providing the opportunity for large-scale representations ., However , the biological interpretation of these representations is limited , as loops represent inconsistencies with trees , but it is unclear how these inconsistencies arose historically , what genomic regions were involved , or how frequently an exchange happened ., Other types of representations , sometimes named explicit networks 13 , 22 , do aim to provide a historical account in terms of a chronology of events ., Ancestral recombination graphs ( ARGs ) provide potential explanations of the observed data in terms of a progression of recombination and mutation events ., As in trees , mutations are represented as events along the branches ., Recombinations , however , appear as the fusion of two parental branches into one offspring branch ., ARGs provide simple histories that can be used in association mapping 23–25 , SNP genotyping 26 or inference of the frequency and scale of recombination 27 ., However , these applications are hindered by the computational infeasibility of constructing ARGs that explain hundreds of sequences ., The construction of minimal ARGs , containing the minimum number of recombination events required to explain the sample in absence of convergent evolution and back-mutation , is an NP-hard problem 28–30 ., Several approximations have been developed in the last few years , including galled trees 31 , 32 , branch and bound 33 , heuristic 23 and sequentially Markov coalescent approaches 34 ., Recently , a new framework to study genomic relationships has been proposed 35–37 , based on topological data analysis 38–40 ., Topology is the area of mathematics that aims to characterize properties of spaces up to continuous deformations , for instance the number of disconnected components , loops and holes of a space ., TDA extends the concepts and tools of topology to finite metric spaces , that is , finite sets of points and distances between them ., Taking the premise that a set of points has been sampled from an unknown underlying space , TDA attempts to infer the topological features of the space ( Fig 1A ) ., Stability results 35 , 41 , 42 guarantee that small fluctuations in the data only create small changes in the inferred topological features , providing robust characterizations of the data ., In a TDA framework , genomes are characterized by points in a high dimensional space where pairwise distances are genetic distances between sequences ., Assuming that each genomic site mutates at most once across the evolutionary history of the sample , the genetic distance between two genomes can only increase with the acquisition of novel mutations ., The only way of “closing” a loop ( a close path ) in this space is therefore by means of a recombination event 35 ., Hence , an approach to studying recombination in the sample of genetic sequences is to study the loops that those sequences generate when represented in the above way ., A valuable attribute of TDA methods is that they are informative about the scale or size of the inferred topological features ., Given a finite set of data points , there is an infinite number of spaces that are compatible with the points ., TDA structures this spectrum of possibilities by introducing a notion of scale ( Fig 1B ) : at a given scale ϵ , two points are connected in the underlying space if their distance is smaller than ϵ ., Topological features compatible with the data can be then summarized in terms of sets of intervals , named barcodes 43 ( Fig 1C ) ., Each interval in a barcode represents the range of scales across which a particular topological feature ( e . g . a loop ) is present in the inferred topological space ., In the genomic context introduced above , barcodes of loops summarize the frequency and scale ( mutational distance between recombining sequences ) of recombination events , and provide a basic structure on which statistics of genomic exchange can be built 37 ., TDA methods are particularly well suited for large datasets ., In the context of molecular phylogenetics and evolution , they have been applied to the study of viral recombination and reassortment 35 , bacterial species 36 and point estimators in population genetics 37 ., However , these implementations of TDA have limitations , as they are not tailored for the biological problem they try to address ., Specifically , traditional TDA methods only use information about genetic distances between sequences , and so they discard the full structure of segregating characters , missing numerous recombination events that are required to explain the data ., Relatedly , it is unclear which specific evolutionary histories explaining the data TDA informs about , and what is the precise relation between barcodes and these histories ., Here we address these two important aspects , improving on the scalable capabilities of TDA to extract robust information on the possible evolutionary histories of a sample of genetic sequences ., In particular , we show that by systematically sampling subsets of segregating sites and performing TDA , we are able to identify most of the necessary recombination events identified by bound methods 33 , 44 , 45 , providing a significant improvement of past methods 35–37 in terms of interpretation and sensitivity ., Moreover , we introduce a novel type of graph ( topological ARG or tARG ) , closely related to minimal ARGs , that captures ensembles of minimal recombination histories; and we show that TDA informs about the topological features and genetic scales of these graphs ., Like minimal ARGs 22 , 23 , tARGs can be considered as explicit , parsimonious , interpretable phylogenetic representations ., The main advantage of tARGs and barcodes versus minimal ARGs is , however , the possibility of obtaining such phylogenetic information in polynomial time , which allows us to deal with hundreds of sequences ., We have implemented this method in a software , called TARGet , and have illustrated it with several examples , including small migration between diverging populations , human recombination , and horizontal evolution of finches inhabiting the Galápagos archipelago ., The software , instructions and example files used in the manuscript can be obtained from https://github . com/RabadanLab/TARGet ., An ARG is an explicit phylogenetic network representing a possible evolutionary history of a sample of genetic sequences , where only mutation and recombination events are present and convergent evolution is not considered and so never occurs 22 , 46 , 47 ., ARGs are very useful constructs in population genetics and phylogenetics ., However , the problem of building a minimal ARG from a set of genetic sequences is known to be NP-hard 28–30 ., The use of ARGs has therefore been traditionally limited to small samples , consisting of a handful of sequences ., In this section , we introduce a particular class of minimal ARGs and a set of related graphs ., Then , using computational algebraic topology , in the next section we show that it is possible to extract , in polynomial time , phylogenetic information from this class of minimal ARGs , without having to explicitly construct them ., Thus , by restricting to this specific class of graphs , we are able to extend the realm of ARGs to large samples of sequences ., To be specific , we consider a sample S consisting of n distinct genetic sequences with m binary segregating characters ., The latter can be single nucleotide polymorphisms ( SNPs ) , indels , gene duplications or any other genetic trait that takes one of two possible states , 0 or 1 , in each sequence ., An ARG is then formally defined as a directed acyclic graph N with n leaf nodes and a unique root node , where every node other than the root has in-degree one ( tree node ) or two ( recombination node ) , every segregating character labels a unique edge in N ( infinite sites assumption ) , and every sequence in S labels a unique leaf in N . Moreover , each node in N is labelled by a m-length binary sequence , such that the sequence labelling a tree node differs from the sequence of the parent node only at the character labelling the edge that connects the two nodes; and the sequence labelling a recombination node is a combination of the sequences labelling the two parent nodes ., Single-crossover recombinant sequences are formed by taking the first k sites from the sequence of one of the parent nodes ( prefix ) and appending the last m − k sites from the sequence of the other parent node ( suffix ) , for k ∈ 1 , m − 1 ., There is an infinite number of ARGs that can explain a given sample S 22 ., A stochastic model , such as the coalescent model with recombination 46 , 48 , would assign probabilities to each possible ARG ., Here , however , we adopt a parsimony approach and consider ARGs that are minimal ( in a sense defined below ) , without assuming an underlying probabilistic model ., Such a model-independent approach has proven useful in summarizing genetic sequences into evolutionary histories where all events are required ., Specifically , we consider ARGs that contain exactly the minimum number Rmin of single-crossover recombinations required to explain the sample , and that minimize the function, D ( N ) = ∑ r = 0 R min d r ( 1 ), where the sum runs over all recombination events in N , and dr is the Hamming distance between the two parental sequences involved in the r-th recombination ., This is a more restricted definition of minimal ARG than the one that usually appears in population genetics literature 22 , where the condition on D ( N ) is generally not required ., We use the term ultra-minimal ARG to refer to this restricted type of minimal ARG ., Ultra-minimal ARGs are thus minimal ARGs where recombination events involve parental sequences that are as genetically close as possible ., They introduce a higher level of parsimony than minimal ARGs , being informative not only about the minimum number of recombination events , but also about the minimum genetic distance between the recombining sequences that took part in those events ., By construction , an ultra-minimal ARG explaining any given sample always exists ., Examples are shown in Figs 2 and 3 ., A minimal ARG can be condensed by collapsing all unlabelled edges , so that the resulting graph can be embedded into an m-dimensional hypercube and its diagonals ( that is , the line segments joining non-consecutive vertices ) ( Fig 2 ) ., The number of edges and vertices of such a condensed representation is m + 2Rmin and m + Rmin + 1 , respectively , whereas the number of independent loops is Rmin , where a loop is said to be independent if it cannot be embedded in the union of other loops ., In this representation , the distance between two nodes is defined as the number of edges in the shortest path connecting the nodes , and is equal to the Hamming distance between the corresponding sequences ., Given a sample S of genetic sequences , we would like to obtain information about the ultra-minimal ARGs that explain S , without explicitly constructing them ., To that end , we consider the undirected graph G = ( V , E ) , with vertices V and edges E = E1 ∪ … ∪ El , that results from the union of all condensed ultra-minimal ARGs G i = ( V , E i ) explaining S and having the same set of vertices V ( Fig 4 ) ., We call this construction topological ARG ( tARG ) ., A tARG therefore summarizes the collection of most parsimonious histories associated to a sample of genetic sequences ., However , unlike minimal ARGs , tARGs are completely determined by their vertices ., By considering tARGs instead of minimal ARGs , we are able to reduce an NP-hard problem into a much simpler ( but still very informative ) topological problem , as we describe in next section ., Topological data analysis has emerged during the last decade as a branch of applied topology that attempts to infer topological features of spaces ( such as the number of loops and holes ) from sets of sampled points 38 ., The topological features of a space are preserved under continuous deformations of the space and can be arranged in mathematical structures called homology groups 49 ., We refer the reader to refs ., 49 , 50 for formal definitions and basic introductions to algebraic topology ., In brief , the nth homology group of a space is an algebraic structure that encompasses all ( n + 1 ) -dimensional holes of the space ., Of special interest to us is the first homology group , whose elements correspond to loops ., Homology groups can be computed by replacing the original space with a simpler one , known as simplicial complex , which has the same topological features as the original space but consists of a finite set of elements ( Fig 1B ) ., A simplicial complex is a generalization of a network that , in addition to nodes and vertices , includes higher dimensional elements like triangles and tetrahedra ., Simplicial complexes are powerful because they allow the implementation of algebraic operations to extract the topological features of the space ., When only a finite set of points of the space is given , there is still a well-defined notion of homology groups , known as persistent homology 39 , 40 , which capture the topological features of the underlying space ., At each value of a scale parameter ϵ , a simplicial complex ( known as Vietoris-Rips complex ) can be constructed by considering the intersections of balls of radius ϵ centred at the sampled points ( Fig 1B ) ., Points are joined if their corresponding balls intersect ., This process produces a sequence of simplicial complexes parametrized by ϵ , from which persistent homology can be computed using available algorithms 39 , 40 ., Remarkably , the computation time of persistent homology is polynomial in the number of points 39 , 40 ., Persistent homology can be represented using barcodes 43 ., These are graphical representations where each element of persistent homology is represented by a segment spanning the interval ϵb , ϵd , where ϵb and ϵd are the values of the parameter ϵ at which the corresponding feature is respectively formed and destroyed in the sequence of simplicial complexes ( Fig 1C ) ., Thus , each segment in a barcode represents a topological feature inferred from the data , and the position and length of the segment are informative of the size of the topological feature ., The values ϵb and ϵd are referred as birth and death time of the topological feature , respectively ., In the current context , we exploit the use of persistent homology to infer topological features of an unknown tARG , given a set of sampled nodes ( Fig 5 ) ., The use of persistent homology to detect the presence of recombination in genetic samples was proposed in 35 ., However , the relation between persistent homology and explicit evolutionary histories incorporating recombination events was not studied ., Our aim is inferring information about the loops of the tARG , as they correspond to recombination events present in the collection of most parsimonious histories explaining the sample ., To that end , we consider the Hamming distance matrix of the sample and compute persistent homology using the algorithm developed in ref ., 39 , 40 ., Since computing the distance matrix and persistent homology requires respectively O ( n 2 m ) and O ( n 3 ) operations 39 , 40 , the running time grows at most cubically with the number of genetic sequences ., An advantage of using persistent homology instead of just counting loops in a nearest neighbour graph is that we also obtain valuable information about the genetic distances between recombining sequences ., The barcode that results from this computation contains information about the number and size of the loops in the tARG underlying the sample ( Fig 5 ) ., Each segment in the barcode represents a loop in the tARG , and therefore a recombination event in an ultra-minimal ARG explaining the sample ., The position of each segment provides information about the genetic scales involved in the corresponding recombination event ., Specifically , 2ϵd sets an upper bound to the mutational distance between the two recombining sequences , since all pairwise distances between nodes in the loop are smaller than 2ϵd ., The number of segments in the barcode ( namely , the dimension of the first persistent homology group ) or persistent first Betti number , b1 , is hence a lower bound of the number of recombination events in the tARG , R ¯ min ., Note that , since a tARG is the union of multiple minimal histories , R ¯ min can be larger than Rmin ., In particular , R ¯ min > R min when there are three characters for which all eight possible allele combinations appear in the sample ., In general , this can only happen at very large recombination rates ., The sensitivity of persistent homology to detect recombination decreases as the number m of segregating characters increases ., Indeed , in that case the dimensionality of the ambient space is larger and the sample becomes sparser ., For this reason , b1 is in general a loose lower bound of R ¯ min ., To address a similar problem , Myers and Griffiths introduced the idea of combining the local bounds that result from partitioning the sequence , building a more stringent global bound 45 ., In this way , information about the ordering of characters is incorporated and the location of recombination breakpoints is constrained in the sequence ., This general idea was applied in 45 to the haplotype bound , n − m − 1 ≤ Rmin , to built a stronger lower bound of Rmin , denoted RMG ., A similar idea can be applied in the context of barcodes to build a barcode ensemble , given by the disjoint union of the persistent first-homology barcodes of a set of optimally chosen , non-overlapping intervals within the sequence alignment ( Fig 6A ) ., Given a partition of a genetic sequence , the barcode associated to each interval captures information about recombination events with breakpoint in that interval ., Due to the curse of dimensionality mentioned in the previous paragraph , the union of the barcodes associated to two contiguous genomic intervals often captures more recombination events than the barcode associated to the union of the two genomic intervals ., Therefore , by systematically exploring all possible partitions of the genetic sequence , it is possible to find a partition that maximizes the total number of bars in the barcodes ., The solution is often not unique , as different partitions may lead to the same total number of bars ., One may reduce this degeneration by considering additional criteria , such as also maximizing the total length of the bars ( so that they are more informative about genetic distances ) ., The formal details of the barcode ensemble construction are presented in the Methods section ., The barcode ensemble incorporates information about the full structure of characters in the sample , largely increasing the sensitivity of persistent homology to recombination and providing information on the location of the recombination breakpoints in the sequence ., The number of bars in the barcode ensemble , b ¯ 1 , is an improved lower bound of R ¯ min , in the same way as RMG is an improved lower bound of Rmin:, tARG → R ¯ min ≥ b ¯ 1 ≥ b 1 ↑ ( ultra ) minimal\xa0ARG → R min ≥ R MG ≥ n − m − 1 In biological data , b ¯ 1 and RMG are in general very close to each other ( Fig 6B ) , as tARGs with R ¯ min > R min occur very rarely ., However , unlike RMG , barcode ensembles provide additional phylogenetic information , such as bounds on the mutational distances between recombining sequences ( note that birth and death times in barcode ensembles refer to local genetic distances , namely mutational distances across the genomic interval associated to the particular bar ) ., These features put barcode ensembles at the very interesting interface between the fast , but phylogenetically limited , existing lower bounds to Rmin; and the slow , but phylogenetically rich methods for reconstructing minimal ARGs ., We have implemented the computation of barcode ensembles in publicly available software , called TARGet ., We consider five examples that illustrate how the formal developments presented in previous sections can be used to extract useful phylogenetic information from samples of genetic sequences ., The first example is a simple toy model where an explicit minimal ARG can be easily constructed ., It displays how the information contained in the barcode ensemble of the sample directly maps to features of ultra-minimal ARGs ., The second example , based on simulated data of two sexually reproducing populations exchanging genetic material at low rate , shows the applicability of persistent homology to large datatsets , consisting of several hundreds of sequences ., It also demonstrates the use of phylogenetic information contained in the barcode ensemble to distinguish among various biological settings with similar recombination rates ., The third and fourth examples consist respectively of 250 and 100 kilobase regions in the HLA and MS32 loci of ∼ 100 humans , where several meiotic recombination hotspots localize ., The fifth example consists of a 9 megabase scaffold in the genome of 112 Darwin’s finches 51 ., These last three examples serve to illustrate the applicability of barcode ensembles to real datasets ., The examples above illustrate the use and interpretation of barcode ensembles in molecular phylogenetics ., As we have discussed , an important feature of topological approaches to phylogenetics is that they inform about most parsimonious evolutionary histories ., Being model-independent approaches , they describe minimal sets of events required to explain a sample of sequences , without assuming any probabilistic model of evolution ., In some situations , however , we are interested in estimating the parameters of a specific evolutionary model from the observed data ( e . g . the recombination rate in a coalescent model with recombination ) ., To that end , barcode ensembles can be taken as summary statistics from which to build parameter estimators ., For instance , in Fig 11A we show the dependence of b ¯ 1 on the recombination rate for a set of 1 , 000 coalescent model simulations ., The expected b ¯ 1 of the barcode ensemble is informative of the recombination rate , growing monotonically with the later ., Compared to sequentially Markov coalescent ( SMC ) approaches for ARG inference 34 , b ¯ 1 is strongly correlated with the number of recombinations in SMC ARGs derived from the same set of sequences ( Pearson’s r = 0 . 93 , p < 10−100 , S1 Fig ) ., Although the coefficient of variation is ∼ 35% larger for b ¯ 1 ( S1 Fig ) , its computing time is substantially lower ( > 9 times faster after parallelizing in a modern 8-cores desktop computer , S1 Fig ) , being a robust approach to coalescent-model recombination rate estimation in large datasets ., Furthermore , unlike the number of recombinations in SMC ARGs , b ¯ 1 is unbiassed at small recombination rates , vanishing when the recombination rate is zero ( Fig 11A ) ., Although recombination rate estimation is a very direct example , the barcode ensemble of a sample of genetic sequences contains other rich phylogenetic information apart from b ¯ 1 , which can be used for more complex parameter estimation in structured models of evolution ., Consider , for instance , the case of two divergent populations with migration and recombination discussed above ., In this model , the average genetic distance between recombining sequences is expected to decrease with the migration rate , as the average time to the most recent common ancestor between foreign and local gametes in a population is shorter ., In Fig 11B we show the dependence of the average death time ( 〈ϵd〉 ) on the migration rate parameter , for the barcode ensembles of a set of 900 coalescent model simulations with fixed recombination and variable migration rates ., As expected , 〈ϵd〉 is informative of the migration rate , decreasing monotonically with the later ., It is therefore a good measure for estimating migration rates ., Consistently , 〈ϵd〉 correlates with time to the most recent common ancestor of recombining sequences in SMC ARGs obtained from the same data ( Pearson’s r = 0 . 55 , p < 10−72 , S1 Fig ) ., Although the coefficient of variation of 〈ϵd〉 is ∼ 60% larger ( S1 Fig ) , extracting this type of information from SMC ARGs requires the implementation of a greedy algorithm , substantially increasing the running time ( ∼ 8 times slower in a single core of modern desktop computer , S1 Fig ) and therefore limiting its applicability to large datasets ., These two simple examples illustrate the utility of barcode ensembles for building parameter estimators in specific models of evolution ., Importantly , being model-independent , they are robust and flexible tools which can be applied in an infinitely large number of possible evolutionary models ., As the famous title of the essay by Dobzhansky “Nothing in Biology Makes Sense Except in the Light of Evolution” underscores , evolutionary processes are central orchestrating themes in biology ., Mutations , recombinations and other evolutionary processes get imprinted into genomes through selection , reflecting the accumulated history giving rise to an organism ., Phylogenetics try to reconstruct the evolutionary history through the comparison of genomes of related organisms ., In addition to reporting relationships and elucidating particular histories , one would like to understand and quantify how different evolutionary processes have occurred ., The identification and quantification of evolutionary processes can be challenging due to the lack of a well-established universal framework to capture evolutionary relationships beyond trees ., In addition , robust statistical inference needs to exploit the large number of genomes that are now becoming available , aggravating the computational burden and obscuring interpretations ., Ideally , we would like to have a biologically interpretable framework able to quantify different evolutionary processes by analyzing large numbers of genomes ., In this paper we have proposed a few steps in this direction ., We have extended the notion of barcodes in persistent homology to identify the genetic scale and number of recombination events ., We have shown that , by correctly studying persistent homology in subsets of segregating sites , it is possible to characterize the genomic regions where recombination takes place and identify the gametes involved in particular recombination events ., The persistent homology barcodes derived from each of these sets can be structured as a “barcode ensemble” where each bar captures a recombination event ., Barcode ensembles can be interpreted as counting and quantifying the scale of recombination events in a variation of Ancestral Recombination Graphs ( ARGs ) ., Topological ARGs represent a summary of potential recombination histories that can explain the data ., The method proposed , TARGet , is scalable to hundreds of genomes ., As an alternative to some phylogenetic networks , barcode ensembles provide robust quantification of events , the distribution of genetic scales , computational scalability and interpretative graphs ., Barcode ensembles are versatile in that they do not assume any specific model of evolution , providing explicit , interpretable summaries of the minimal set of recombination events required to explain a sample of genetic sequences ., Here we have illustrated their use in several practical cases ., However , the range of possible applications is unlimited ., In some cases , it may be convenient to perform minor modifications to the approach described here ., For instance , although in our exposition we have only made use of Hamming distance and binary sequences , the main concepts we have presented extend straightforwardly to other genetic distances ., The use of these metrics can be particularly useful in cases with rapidly diverging samples or substantial mutational biases ., In other cases , information about the ancestral and derived alleles for each character in the sample may be available ., Although tARGs have no natural directionality , the inclusion of the ancestral sequence in the original sample may lead in those cases to more stringent bounds on R ¯ min , similarly to what occurs with other approaches to recombination inference 22 ., Finally , more efficient integer linear programming algorithms , like the one of 33 , could in principle be also generalized to the computation of barcode ensembles ., We extended the construction of ref ., 45 to persistent homology barcodes ., From a geometric perspective , this corresponds to projecting the original space on sets of mutually orthogonal hyperplanes in the ambient hypercube , and computing persistent homology in each of those projections ., For that aim , we need to establish an ordering relation on barcodes ., Being sets of intervals , it is natural to take the maximum of two barcodes to be given by the one with largest L0-norm , namely largest b1 ., If both barcodes have the same L0-norm , we may successively compare other norms ( e . g . other Lp-norms ) , until the tie is broken or , otherwise , one of the two barcodes is arbitrarily chosen ., The algorithm of 45 is then generalized to persistent homology barcodes as follows: The barcode ensemble of S is the union barcode R 1 m that results from this algorithm ., We implemented the algorithm in a publicly available multi-threaded software , TARGet , which is distributed under the GNU General Public License ( GPL v3 ) ., The application is fully written in Python 2 . 7 , and relies on Dionysus C++ library for persistent homology computations ( http://www . mrzv . org/software/dionysus ) ., Since considering all

Introduction, Results, Discussion, Methods

The recent explosion of genomic data has underscored the need for interpretable and comprehensive analyses that can capture complex phylogenetic relationships within and across species ., Recombination , reassortment and horizontal gene transfer constitute examples of pervasive biological phenomena that cannot be captured by tree-like representations ., Starting from hundreds of genomes , we are interested in the reconstruction of potential evolutionary histories leading to the observed data ., Ancestral recombination graphs represent potential histories that explicitly accommodate recombination and mutation events across orthologous genomes ., However , they are computationally costly to reconstruct , usually being infeasible for more than few tens of genomes ., Recently , Topological Data Analysis ( TDA ) methods have been proposed as robust and scalable methods that can capture the genetic scale and frequency of recombination ., We build upon previous TDA developments for detecting and quantifying recombination , and present a novel framework that can be applied to hundreds of genomes and can be interpreted in terms of minimal histories of mutation and recombination events , quantifying the scales and identifying the genomic locations of recombinations ., We implement this framework in a software package , called TARGet , and apply it to several examples , including small migration between different populations , human recombination , and horizontal evolution in finches inhabiting the Galápagos Islands .

Evolution occurs through different mechanisms , including point mutations , gene duplication , horizontal gene transfer , and recombinations ., Some of these mechanisms cannot be captured by tree graphs ., We present a framework , based on the mathematical tools of computational topology , that can explicitly accommodate both recombination and mutation events across the evolutionary history of a sample of genomic sequences ., This approach generates a new type of summary graph and algebraic structures that provide quantitative information on the evolutionary scale and frequency of recombination events ., The accompanying software , TARGet , is applied to several examples , including migration between sexually-reproducing populations , human recombination , and recombination in Darwin’s finches .

infographics, taxonomy, genome evolution, phylogenetics, data management, mathematics, algebra, genome analysis, dna, homologous recombination, computer and information sciences, genomics, molecular evolution, evolutionary systematics, evolutionary genetics, biochemistry, data visualization, nucleic acids, graphs, genetics, algebraic topology, biology and life sciences, topology, physical sciences, dna recombination, evolutionary biology, computational biology

journal.pcbi.1003503

Modeling Mutual Exclusivity of Cancer Mutations

Recent years in cancer research are characterized by both accumulation of data and growing awareness of its overwhelming complexity ., While consortia like The Cancer Genome Atlas ( TCGA ) 1 and the International Cancer Genome Consortium ( ICGC ) generate the multidimensional profiles of genomic changes in various cancer types , computational approaches struggle to pinpoint its underlying mechanisms 2 ., The most basic yet already challenging task is to identify cancer drivers , genomic events that are causal for disease progression ., A second , more general task is to elucidate sets of functionally related drivers , such as mutations of genes involved in a common oncogenic pathway ., One systematic approach to address the latter task is to search for mutually exclusive patterns in cancer genomic data 3–7 ., Typically , the data is collected for a large number of tumor samples , and records presence or absence of genomic alterations , such as somatic point mutations , amplifications , or deletions of genes ., In mutually exclusive patterns , the alterations tend not to occur together in the same patient ., These patterns are commonly characterized by their coverage and impurity ., Coverage is defined as the number of patient samples in which at least one alteration occurred , while impurity refers to non-exclusive , additional alterations ( referred to as non-exclusivity or coverage overlap in previous studies ) ., Such mutually exclusive alterations have frequently been observed in cancer data 8–10 and were associated with functional pathways or synthetic lethality 3–8 , 11 , 12 ., Therefore , mutually exclusive patterns are important for a basic understanding of cancer progression and may suggest genes for targeted treatment ., Previous studies identified mutually exclusive patterns either via integrated analysis of known cellular interactions and genomic alteration data 6 , or de novo , by an online learning approach 3 , or by maximizing the mutual exclusivity weight introduced by Vandin and colleagues 4 , 5 , 7 ., The weight increases with coverage and decreases with coverage overlap 4 and proved successful for pattern ranking and cancer pathway identification ., To our knowledge , there exists no approach that explicitly models the generative process of mutual exclusivity patterns ., In the absence of a statistical model of the data , the definition of the weight , although intuitively reasonable , remains arbitrary ., In the previous studies , the weight served also as statistic for a column-wise permutation test that assesses the significance of patterns ., We show that the power of this test decreases with the number of genes , likely because the weight does not scale with gene number , and the same impurity level affects it more with more genes in the pattern ., Most importantly , none of the existing approaches deal with the problem of errors in the data ., Despite advanced methodologies on both experimental and computational side 13 , records of genomic alterations may contain false positives and false negatives , due to measurement noise , as well as uncertainty in mutation calling and interpretation ., As illustrated in Figure S1 , ignoring errors in the data , particularly false positives , may lead to wrong ranking of patterns ., Here , we develop two alternative models for cancer alteration data ( Figure 1 ) ., One is a probabilistic , generative model of mutually exclusive patterns in the data ., The model contains coverage as well as impurity as parameters , together with false positive and false negative rates ., We show analytically that the model parameters are identifiable , and propose how they can be estimated and used for pattern evaluation ., The second is a null model assuming independent alterations of genes ., Via comparison of the mutual exclusivity model to the null model , our approach allows statistical testing for mutual exclusivity , both in the presence and absence of errors ., First , we evaluate performance of our approach in the case when , as it is done in the literature , the data is assumed to record no false positive or negative alterations ., On simulated patterns our mutual exclusivity test proves more powerful than the weight-based permutation test ., In glioblastoma multiforme data 14 , analyzed by the previous approaches , we find novel , biologically relevant patterns , which are not detected by the permutation test ., Next , we examine the bias introduced in pattern ranking by ignorance of errors , especially false positives , and show that when the error rates are known , our approach is able to accurately estimate the true coverage and impurity and rank the patterns accordingly ., Finally , we analyze the practical limits of accurate parameter estimation in the most difficult , but also most realistic case where the data contains errors occurring at unknown rates ., We apply our approach to a large , pan-cancer collection of 3299 tumor samples from twelve tumor types 15 , for which the model accounting for the presence of false positives can accurately be estimated ., This model is shown to be more flexible than the model assuming no errors in the data , and is applied to identify several universal , significant mutual exclusivity patterns , which would not be found by the previous methods ., A mutual exclusivity pattern can be detected in a given cancer alteration dataset , with columns that correspond to a subset of measured genes and rows ( observations ) that correspond to patients whose tumor samples were collected ( with ) ., For each patient and gene , the dataset records a binary alteration status of the gene observed in the patient , with 0 standing for absence and 1 for presence of alteration ., We assume that the mutual exclusivity patterns are the result of the following generative process ( Figure 1A ) ., First , with a certain probability , denoted and called coverage , the patients who are covered by the pattern are chosen ., Each row corresponding to a covered patient is hit by an exclusive alteration , meaning that exactly one gene is assigned value 1 in this row ., Here , we assume that all genes have equal probability to be exclusively mutated ., Next , in the same row , with probability , any other gene can be mutated in addition ., Those added alterations are interpreted as impurity in the mutual exclusivity pattern , hence is referred to as the impurity parameter ., The generative process described up to this point coincides with the data simulation procedure used in previous studies 4 , 5 ., However , the corresponding generative model was not used for statistical inference ., This prevalent view of the generative process ignores the possible occurrence of errors ., Realistically , the observed alteration data result from adding false positives ( with rate ) and false negatives ( rate ) to the true , exclusive , and impure alterations ., We propose a generative model of mutual exclusivity that describes the process illustrated in Figure 1A ., For each patient in a given dataset , the proposed model ( Figure 1B and Methods ) assigns a probability to the corresponding observation ., The model is defined by a set of hidden random variables , , , and observed variables ., The binary variable has value 1 with probability , indicating that the patient is covered by the mutual exclusivity pattern ., The hidden random variable points at the gene that is exclusively altered in that pattern ., The set of hidden random binary variables corresponds to the true alteration status of the genes , and the set of observed binary variables corresponds to the alteration status that is recorded in the data ., Each true alteration variable has value 1 either if it was chosen to be exclusively altered , or if it was not chosen but acquired an impure alteration with probability ., The values of the variables are the same as values of , except for cases of false positives ( with probability ) and false negatives ( with probability ) ., First , we analyzed the identifiability of the model from observed data ( Text S1 ) : Proposition 1 For , the parameters in the mutual exclusivity model are identifiable ., Encouraged by this result , we propose an expectation maximization algorithm ( Methods ) to estimate the maximum likelihood parameter values and evaluate its performance in practice ( Results ) ., In the case when the dataset does not carry the mutual exclusivity pattern , we assume that the corresponding genes are mutated independently with their individual alteration frequencies ., This is modeled with a set of independent , observed binary random variables , satisfying for each ( referred to as the independence model; Text S1 ) ., We devise a mutual exclusivity test ( shortly , ME test ) , which compares the likelihood in the mutual exclusivity model to the likelihood in the independence model ., Since the models are not nested , we use Vuongs closeness test 16 to compute the p-values ( Methods ) ., A small p-value means that the mutual exclusivity model is closer ( with respect to Kullback-Leibler divergence ) to the true model from which the data was generated than the independence model ., The test statistic accounts for the difference in degrees of freedom between the models ., We evaluate our mutual exclusivity model and statistical test in three different scenarios ., First , we make an assumption prevalent in the literature , namely that the data is generated without errors ., In the second scenario , we assume that the data contains errors , and the error rates are given ., Finally , we consider the scenario where the data is generated with errors , and the error rates are unknown ., First , we evaluate the performance of our mutual exclusivity model on simulated data assuming that the data is clean of errors ., In this case , the model is reduced , since it is parametrized only by the coverage and the impurity , and the observed variables are equated with the true hidden variables ., We have derived closed-form expressions for the maximum likelihood parameter values ( Methods ) , providing reliable parameter estimates already for datasets of sample size 200 ( Table S1 ) ., We simulated datasets from the reduced mutual exclusivity model , for increasing gene set sizes , , patients , and combinations of parameter values and , with 20 datasets generated per each parameter setting ( example in Figure 2A ) ., For each dataset , we assessed the significance of mutual exclusivity using the proposed ME test ( Methods ) ., For comparison , we obtained empirical p-values from the weight-based permutation test , which permutes individual columns in the dataset 1000 times , and reports the number of times a permuted dataset had a higher weight than the original 4 , 5 ., For datasets with three genes only and low coverages , both our ME and the permutation test not always detect mutual exclusivity ( Figure 2B ) ., As the gene set size increases , in contrast to the permutation test , the ME test becomes more powerful ., With ten genes , our test supports mutual exclusivity for all datasets , whereas the permutation test does not , even for a large fraction of datasets with high coverage ., As an example , for the mutual exclusivity pattern in Figure 2A the ME test p-value is , and the permutation test p-value is 0 . 15 ., We speculate that the reason for the decreased power of the permutation test is the weight itself ., With the same coverage and impurity , large gene sets get less significant weights than small gene sets , since the weight decreases drastically with addition of impure alterations in each row , and this addition is more likely for longer rows ., In addition , with increased gene set size the ME test p-values tend to decrease ., This suggests that the test will remain powerful also after multiple hypothesis testing correction , which is expected to be more restrictive for larger set sizes ., Both tests correctly do not support mutual exclusivity for datasets generated from the independence model ( Figure 2C ) ., 20 datasets were simulated per each maximum individual frequency ( each frequency was drawn at random uniformly from interval ) ., The same , correct behavior was observed when the independent frequencies were drawn from a distribution observed in real cancer data ( Figure S2 ) ., Figures 2B , C show that the ME test , without computationally expensive permutations , yields ranges of p-values that are amenable to multiple testing corrections ., In summary , the ME test is equally powerful for small gene sets as the permutation test , and more powerful for larger ones , and can efficiently be applied in practice ., We further use our model to identify significant mutual exclusivity patterns with high coverage and low impurity in glioblastoma multiforme samples from The Cancer Genome Atlas ( TCGA 14; extended collection; originally published with fewer samples 1 ) ., The data were organized in a binary matrix combining point mutations and copy number variants for 236 patients in 83 genes ., The genes and their alterations were selected to represent significant players and events in disease progression ( Methods ) ., To obtain a comprehensive picture of the types of patterns that can be found in this dataset , we restricted the gene set size to four , and evaluated all 1 , 837 , 620 possible gene subsets of this size ., Figure 3A presents the pattern with the largest weight , but also large imbalance: in that pattern , almost the entire coverage comes from alterations of a single gene , EGFR ., With our approach the quality of each pattern can be assessed with the estimated coverage and impurity parameters , while its significance is given by the p-value from the ME test ., In the standard understanding , a high quality pattern has high coverage and low impurity ., For the GBM dataset we obtained 11 significant ( Benjamini-Hochberg adjusted ME p-value ) patterns with estimated coverage larger than 0 . 3 and impurity lower than 0 . 2 ( Table S2 ) ., Figure 3B–D presents top three of those patterns with the lowest impurity ., Out of the genes included in those top sets , NF1 , PIK3C2G , PIK3R1 and PIK3CA play roles in the interconnected canonical glioblastoma signaling 1 , although are not found directly grouped into individual pathways as identified by the original publication ., Notably , the TRAT1 protein is a known interaction partner of PIK3R 17 , 18 ., Table 1 summarizes the statistics for all presented patterns , underlining the differences between the ME and permutation tests ., With the explicit account for coverage and impurity as parameters in the model , our approach gives control over which important features of the patterns should be used to prioritize the significant patterns of interest ., In contrast to the permutation test , the ME test is specifically designed to prefer balanced patterns ., Consequently , patterns identified using our ME approach have over three times lower median imbalance than the median imbalance of top weight patterns with adjusted permutation test p-values ( Figure S3 ) ., To assess the imbalance of a given gene set , we calculated the ratio between the number of alterations of the gene with the largest individual frequency in the set to the total number of patients covered with the pattern ., Our analysis did not rediscover four mutually exclusive gene sets ( Table S3 ) identified previously based on optimizing the weight 4 , 7 for the first , original GBM dataset version ., Several genes in those sets did not pass our filtering criteria in the pre-processing step ( Methods ) , and one gene set could not be analyzed for this reason ., Two sets had large estimated impurity ( ) , which does not satisfy our threshold ., All three analyzed gene sets were insignificant according to the ME test , most likely due to relatively high imbalance ( two to three times larger than median imbalance of gene sets we identified , compare Figure S3 ) ., Interestingly , one of those gene sets does not have a significant permutation p-value , which may be due to the fact that the processing of the data was different and the original dataset contained fewer samples ., In this section , we consider the scenario where the data are erroneous , and the error rates are known and can be used for pattern evaluation ., Figure S1 visualizes the severe effects of error ignorance ., The observed weight , computed on datasets with false negatives , is consistently reduced as compared to the true weight of patterns generated without errors ., Addition of false positives introduces most bias in the observed weight , and results in false ranking ., Similarly , for the reduced mutual exclusivity model assuming no errors , parameter estimation fails in the case when they do occur in the data ( Figure S4 ) ., Thus there is a well motivated need for the model to account for errors ., Fixing the parameters and in our model to the true false positive and negative rates , respectively , we can estimate the remaining coverage and impurity parameters using the EM algorithm ( Methods ) ., This estimation is very precise for simulated datasets with five genes , and sample sizes 200 or 1000 ( Table S1 , Figure S5 ) ., Figure 4 shows that such precise estimates can be used to rank the patterns by their estimated true quality , first sorting by the estimated impurity and second by their estimated coverage ., We ranked the erroneous datasets simulated in Figure S1 by their estimated true quality ., Next , we evaluated the fraction of dataset pairs which were ordered the same way as when their true impurity and coverage were used for sorting ., This fraction of correctly ranked pairs was compared to the fraction that is ranked the same way by the observed weight as compared to the true weight ., For data containing false negatives both the quality ranking and the observed weight perform very well in correct ranking ., The estimated true quality significantly outperforms the observed weight in the presence of false positives ., Finally , we consider the scenario , where the observed data contains errors that occur at unknown rates ., In this case we need to estimate all four model parameters , and we proved the model to be identifiable from the data ( Text S1 ) ., As expected , Table S1 shows that for realistic sample and gene set size ( 200 or 1000 patients and five genes ) , and for typical parameter settings ( with small impurity and error rates and ) , parameter estimation is more difficult than in the case where and are given ( compare Figure S5 ) ., The estimated parameter values start approaching the true ones only for prohibitively large sample sizes ( Figure S6 ) ., In particular , for realistic sample numbers , the parameter is largely underestimated ., Since in case of mutual exclusivity and small values , there are in total not many true positive cases , the actual false negatives should be very rare ., Thus , without much loss of generality of our approach for realistic datasets , we further assume that the false negative rate is zero , and account only for the false positives ., With this assumption , our approach is still very useful in mutual exclusivity analysis: Figure S1 and Figure 4 show that in terms of ranking there is a pressing need to account for the false positives rather than for false negatives ., Table S1 and Figure S7 illustrate that with this assumption , already for 1000 samples ( but not 200 ) a much more accurate estimation of the remaining parameters , , and is possible ., Still , for impurity too similar to false positive , the parameter is overestimated , and underestimated ., Thus , in some cases , the true impurity may be smaller than its estimated value , making our evaluation of patterns over-conservative ., Again , this problem diminishes for larger datasets ., Figure 5 shows , that for realistic dataset sizes and parameter sizes , the ME test is able to detect mutual exclusivity in data with false positives , and is more powerful than the permutation test ., We applied our approach accounting for false positives to pan-cancer genomic alteration data 15 , a data collection from twelve distinct cancer types ., Combining cancer datasets enables to mine for mutually exclusive patterns that are universal for the disease , but can be a problem for the search of patterns that are specific for one of the combined types ., A gene set which has mutually exclusive alterations in only one cancer type and not others will most likely not be detectable in the combined dataset ., The pan-cancer dataset is much larger than the glioblastoma data , thus allowing more accurate parameter estimation ., Somatic point mutations , copy number variants , and methylations were compiled into a single binary data matrix ., Duplicated columns from the compiled matrix were removed , yielding a matrix with 428 columns , some of which represent not one , but several genes ( Methods ) ., We aimed to collect universal , low-impurity mutual exclusivity patterns for gene sets of size five that cover multiple cancer samples , accounting for possible false positives ., We first pre-filtered the immense set of all possible subsets , starting with fitting the reduced model ( assuming no errors in the data ) for all 15 , 504 subsets of 20 measured genes that were selected by their large individual alteration frequency ( ; c . a . 0 . 6% ) ., Next , we chose the 2039 subsets that had estimated coverage larger than 0 . 3 , impurity lower than 0 . 2 , and ME statistic larger than 0 , indicating the reduced mutual exclusivity model fits the data better than the independence model ( not necessarily significantly ) ., Figure 3E shows the pattern that in the pre-filtered dataset has the largest weight , which is largely dominated by alterations of TP53 ., Finally , we applied the model accounting for false positives to the pre-filtered subsets , and identified 476 high quality patterns ( Table S5 ) with estimated coverage larger than 0 . 3 , impurity lower than 0 . 2 , selecting by significance ( Benjamini-Hochberg adjusted ME p-value ) , and sorting by impurity ( lowest on top; examples in Figures 3F–H ) ., Three out of all columns in the visualized patterns correspond not to one , but a set of genes , and are denoted META 1-3 ( see Table S4 for individual genes ) ., A possible reason for a large number of significant and high quality gene sets ( Table S5 ) is the fact that the identified gene sets overlap ., Such overlapping gene sets may either share strongly mutually exclusive subsets of smaller size , or may all be subsets of a single , larger mutually exclusive gene set ., Findings for various cancers for pairs of genes support that the top patterns are indicative of coexistence in a common cancer pathway ., For instance , for the pattern in Figure 3G , the protein products of the genes PTEN and MYC ( element of META 2 ) are co-regulators of p53 in control of differentiation , self-renewal , and transformation in glioblastoma 19 ., The gene copy ratio of MYC and CDKN2A in the same pattern has a prognostic value in squamous cell carcinoma of the head and neck 20 ., Finally , PTEN and VHL are both known regulators of the HIF-1 pathway 21 ., PTEN and APC , common to two identified gene sets , are tumor suppressors that are known to interact in cancer 22 ., Table S6 compares the p-values and estimated parameters , obtained for the top identified patterns , using the model accounting for false positives to the reduced model ., As a rule , the former p-values are smaller , while the values of the coverage and impurity parameters estimated by the two models are similar ., In one case however ( Figure 3G ) , the estimated false positive rate is 0 . 037 , yielding the estimated coverage accordingly smaller ( 0 . 45 ) than the estimate from the reduced model ( 0 . 55 ) ., This is why this pattern , although with larger observed coverage , in our true quality ranking would score lower than the pattern in Figure 3H ., In general , for all pre-filtered subsets the ME test based on the model that accounts for false positives was more flexible , and returned a larger number of significant p-values ( 1397; adjusted ME p-value ) , than the test based on the reduced model ( 1171 ) ., This work brings two main contributions ., First , a probabilistic , generative model of mutual exclusivity , with readily interpretable parameters that represent pattern coverage and impurity , as well as parameters that account for false positive and false negative rates ., In the case when the data is clear of errors , we give closed-form expressions for maximum likelihood coverage and impurity estimates ., For erroneous data , we propose an EM algorithm for parameter estimation ., We prove analytically that the model parameters are identifiable , and show the limits of parameter estimation in practice , where the sample sizes are small ., These limits allow accurate estimation of the most troublesome false positive rate , as well as the coverage and impurity parameters , which are most useful for pattern ranking ., Second , we develop the ME test , which assesses the significance of mutual exclusivity patterns by comparing the likelihood of the dataset under the mutual exclusivity model to the null model assuming independent alterations of genes ., The proposed test proves to be more powerful than a permutation test applied previously ., Our approach was first applied to identify mutually exclusive patterns that are specific for glioblastoma , with the assumption prevalent in the literature that the data does not contain errors ., The genes that show the top identified patterns are involved in canonical glioblastoma signaling pathways , with addition of two novel genes , RPL5 and TRAT1 ., Next , we applied the model that accounts for false positives , and detected universal patterns with high coverage and low impurity , found significant by the ME test across a collection of samples from twelve different cancers ., Although both these cancer cohorts were already analyzed in detail with cutting-edge tools 1 , 3–7 , 15 , our new testing procedure provides new , significant , and biologically relevant patterns that were not identified previously ., The proposed mutual exclusivity model could be extended in several ways ., For instance , the current model explicitly assumes that the mutually exclusive mutations occur equally likely in all genes in the dataset ., This assumption has two important advantages ., First , the ME test finds most evidence for mutual exclusivity for balanced patterns , where the genes contribute similarly to the coverage ., Second , with this assumption our EM algorithm is very efficient ( Methods ) and dropping it would increase its time complexity ., The model may be extended to allow different mutually exclusive mutation rates of genes as parameters , which would be estimated from the data ., Another possible extension of the model would allow for multiple gene sets , each with own coverage and impurity parameters , and the same error rates ., Such a model , in contrast to previous work in this direction 7 , would correct for errors and prioritize patterns with balanced mutually exclusive mutations ., Finally , this work , focusing on modeling , evaluation , and testing for mutual exclusivity , does not deal with efficient search for mutual exclusivity patterns ., Instead , we browse all possible , small gene subsets measured in glioblastoma , or all gene sets with high coverage in the pan-cancer data ., Integration of the model into existing 4 , 5 or a new search procedure is one direction of our future research ., Ideally , the objective optimized in the search would be a single measure that reflects preferred impurity , coverage , and significance in the ME test ., These three evaluation criteria could be combined using appropriate priors in the ME model ., The results presented here indicate that already now , the proposed approach is a step forward in the demanding task of mining cancer genomic data for the mechanistic principles of this disease ., The TCGA provisional glioblastoma data for 236 patients in 83 genes includes somatic point mutations ( identified as significant by MutSig 23 ) , amplifications and deletions ( called by GISTIC 24 ) ., The combined analyzed dataset is filled with zeros , and has entry 1 whenever there was a significant point mutation , or a copy number variant that is concordant with expression in the data ., For each gene , concordance of its copy number variants ( amplifications and deletions ) with expression data was assessed using the Wilcoxon test , comparing medians of the gene expression in the samples with the variant to expression in diploid samples ., Specifically , amplifications were tested to have expression median higher , and deletions to have the median lower than the diploid cases ., Only significantly concordant ( p-value 0 . 05 ) variants were recorded in the analyzed dataset ., The pan-cancer TCGA data has 3299 samples and records somatic point mutations , amplifications , deletions and methylations ., Pre-processed data was downloaded from the cBioPortal 25 and combined into a single binary matrix with altered genes as columns , separately for the GBM and for the pan-cancer data collection ., In the combined pan-cancer matrix some columns were identical , with different genes having alterations in exactly the same patients ., Since such genes are undistinguishable with respect to mutual exclusivity patterns , they were combined into “meta” sets of genes , and represented with a single column in the matrix ., Let be the set of model parameters , with coverage , impurity , false positive rate and false negative rate ., We define the mutual exclusivity model on a set of random variables: hidden binary random variable that indicates patient coverage , hidden binary vector variable that specifies the single exclusively mutated gene in a covered patient , a set of hidden binary variables that represent the true alterations of genes , and a set of observed variables that correspond to the alteration status of genes recorded in the data ., The model is defined by:for all , where is a unit vector of length with a single entry 1 at position ., Thus , means that some other gene than is selected as mutually exclusively mutated ., With this distribution of , our model is tailored for balanced patterns , where the mutually exclusive alterations occur on average equally frequently for each gene in the pattern ., The set of of hidden binary random variables indicates true alterations in the genes ., has value 1 either when gene is selected as mutually exclusive ( for ) , or , otherwise , when the entry for gene is impure , and it was mutated in addition to another gene ( for ) ., In this model , the observed likelihood for a given observation depends only on the number of values 1 in the observation , denoted , and observation length , and is thus denoted ( Text S1 ) ., For we have: ( 1 ) The likelihood of the whole dataset reads: ( 2 ) where is the number of observations with values 1 in ., Thus , after pre-computation of values in steps , the likelihood can be computed efficiently in only steps of constant time complexity ., For a given set of genes , the mutual exclusivity weight 4 is defined aswhere is the number of samples with at least one alteration in ., To assess significance of the weight , a permutation test is performed with the weight as test statistic , and the null distribution is obtained by independently permuting alterations 1000 times for each gene ( each column in the dataset ) , preserving its alteration frequency ., A summary of this paper appears in the proceedings of the RECOMB 2014 conference , April 2–5 26 .

Introduction, Results, Discussion, Materials and Methods

In large collections of tumor samples , it has been observed that sets of genes that are commonly involved in the same cancer pathways tend not to occur mutated together in the same patient ., Such gene sets form mutually exclusive patterns of gene alterations in cancer genomic data ., Computational approaches that detect mutually exclusive gene sets , rank and test candidate alteration patterns by rewarding the number of samples the pattern covers and by punishing its impurity , i . e . , additional alterations that violate strict mutual exclusivity ., However , the extant approaches do not account for possible observation errors ., In practice , false negatives and especially false positives can severely bias evaluation and ranking of alteration patterns ., To address these limitations , we develop a fully probabilistic , generative model of mutual exclusivity , explicitly taking coverage , impurity , as well as error rates into account , and devise efficient algorithms for parameter estimation and pattern ranking ., Based on this model , we derive a statistical test of mutual exclusivity by comparing its likelihood to the null model that assumes independent gene alterations ., Using extensive simulations , the new test is shown to be more powerful than a permutation test applied previously ., When applied to detect mutual exclusivity patterns in glioblastoma and in pan-cancer data from twelve tumor types , we identify several significant patterns that are biologically relevant , most of which would not be detected by previous approaches ., Our statistical modeling framework of mutual exclusivity provides increased flexibility and power to detect cancer pathways from genomic alteration data in the presence of noise ., A summary of this paper appears in the proceedings of the RECOMB 2014 conference , April 2–5 .

Tumor DNA carries multiple alterations , including somatic point mutations , amplifications , and deletions ., It is challenging to identify the disease-causing alterations from the plethora of random ones , and to delineate their functional relations and involvement in common pathways ., One solution for this task is inspired by the observation that genes from the same cancer pathway tend not to be altered together in each patient , and thus form patterns of mutually exclusive alterations across patients ., Mutual exclusivity may arise , because alteration of only one pathway component is sufficient to deregulate the entire process ., Detecting such patterns is an important step in de novo identification of cancerous pathways and potential treatment targets ., However , the task is complicated by errors in the data , due to measurement noise , false mutation calls and their misinterpretation ., Here , we propose a fully probabilistic , generative model of mutually exclusive patterns accounting for observation errors , with interpretable parameters that allow proper evaluation of patterns , free of error bias ., Within our statistical framework , we develop efficient algorithms for parameter estimation and pattern ranking , together with a statistical test for mutual exclusivity , providing more flexibility and power than procedures applied previously .

systems biology, mathematics, statistics (mathematics), genome analysis, genetics, statistical methods, biology and life sciences, physical sciences, genomics, computational biology

journal.pcbi.1002700

Multistationary and Oscillatory Modes of Free Radicals Generation by the Mitochondrial Respiratory Chain Revealed by a Bifurcation Analysis

The electron transport chain links the central carbohydrate energy metabolism with ATP synthesis ( see Fig . 1 ) ., It transforms the free energy released by the oxidation of NADH and succinate into a form of transmembrane electrochemical potential ( ΔΨ ) , which is used for ATP synthesis 1 ., Reactive oxygen species ( ROS ) are byproducts of electron transport 2 ., They play the roles of both metabolic signals and destructive agents 3–8 ., These key roles of electron transport in cellular metabolism motivate the great interest in understanding the details of the dynamics of this process ., Electron flow through the chain of carriers is controlled by levels of substrates ( NADH , succinate ) , levels of tricarboxylic acid ( TCA ) cycle intermediates , the rate of ATP consumption , oxygen availability , etc 9 ., However , many interesting dynamical properties of the electron transport and linked ROS production are determined by the intrinsic properties of the electron transport chain , such as the structural and functional links between carriers ( topology of the system ) and values of parameters , e . g . reaction rate constants ., Such intrinsic properties can determine physiologically important modes of respiratory chain operation and how transitions between these modes occur ., This relationship between intrinsic properties and dynamics can be understood by analyzing a detailed model of electron transport ., We have reported elsewhere that the Q-cycle mechanism of electron transport in respiratory complex III exhibits bistability 10 , i . e . two stable steady states may exist under the same microenvironmental conditions ( corresponding to two stable steady state solutions of a system of ordinary differential equations ( ODEs ) at the same parameter values ) ., The importance of bistability resides , in particular , in the fact that it can be a main determinant of the destructive effects of hypoxia/reoxygenation 3 , 11 ., Bistability also occurs in a model of the whole respiratory chain that agrees quantitatively with the measured forward and backwards fluxes through the respiratory chain 12 ., Current experimental techniques make it possible to monitor the behavior of a single mitochondrion in living cells 13 ., This has provided evidence that mitochondria can operate in numerous qualitatively distinct modes ., They can persist in a steady state characterized by a high value of ΔΨ and a low rate of ROS production , can switch to another steady state characterized by a low value of ΔΨ and a high rate of ROS production , and can also enter into a mode of sustained oscillations 13–15 ., The two qualitatively different steady states can be associated with the normal physiological state and a pathological one that can be approached after hypoxia/reoxygenation ., The oscillatory behavior is probably very important for intracellular signaling , as it was found for Ca2+ signaling 16 ., An application of a systematic method that reveals qualitatively different modes of model behavior and corresponding regions of relevant problem parameters , i . e . a bifurcation analysis of a mathematical model describing mitochondrial electron transport , would give insight into these physiologically important modes of mitochondrial functioning and the mechanism of switching between modes ., The objective of the present study is to advance in this direction ., This paper presents a bifurcation analysis of four increasingly complicated models ., The first two of these describe only complex III ( Fig . 2 ) , in forms simplified compared with those previously presented 10 ., The last two include elements of the respiratory chain , shown in Fig . 1 , that were previously modeled in 12 ., For respiratory complex III , the models assume that the core of the complex contains four redox sites: cytochrome b with its two hemes , bH and bL , cytochrome c1 , and the Rieske protein containing iron-sulfur center ( bH-bL-c1-FeS ) ., In addition , the core can bind a two-electron carrier ubiquinone either in the matrix ( Qi-Qi- ) or cytosolic ( -Qo-Qo ) side of the inner mitochondrial membrane ., Symbols are repeated to represent the two electrons ., This gives four different configurations of complex III: ( bH-bL-c1-FeS , bH-bL-c1-FeS-Qo-Qo , Qi-Qi-bH-bL-c1-FeS , Qi-Qi-bH-bL-c1-FeS-Qo-Qo ) ., The models take into account that respiratory complexes constituting the respiratory chain consist of a number of fixed in space electron carriers that can be either reduced ( red ) or oxidized ( ox ) ., A possible state of a complex is defined by a combination of redox states of individual carriers constituting it ., The model variables correspond to these states ., The repeated symbols each correspond to three possible states: either with two valence electrons and two corresponding protons ( ubiquinol ) , one valence electron ( semiquinone ) , or no valence electrons ( ubiquinol ) ., Thus the configuration Qi-Qi-bH-bL-c1-FeS-Qo-Qo has 144 possible states , each corresponding to a variable in the model ., These large numbers of variables make it difficult even to write down explicitly the corresponding systems of ordinary differential equations ( ODEs ) ., Therefore an algorithm is designed and implemented to automatically compute the values of the right hand sides of the ODE system at each step of a numerical solution of a corresponding initial value problem ( IVP ) , based on the rules formulated in accordance with the reaction mechanism 10 ., Two complimentary techniques are combined for numerical bifurcation analysis to systematically search for various types of model behavior and for intervals of parameter values corresponding to multiple steady state solutions or oscillatory solutions:, 1 ) using IVP solvers to solve IVPs for our ODE systems , and, 2 ) using the numerical bifurcation analysis software CL-MATCONTL 17 , 18 ( Text S1 ) to study the corresponding large equilibrium systems ., A numerical bifurcation analysis of the respiratory chain was first conducted for a model of complex III simplified to 145 equations ( referred to further as model 145 ) as described in Methods , with basic set of parameters listed in Table 1 ., This model accounts for only one configuration of complex III , namely the core ( consisting of cytochrome b with its two hemes bH and bL , cytochrome c1 , and the Rieske iron-sulfur ( FeS ) center ) together with ubiquinone molecules bound at both inner ( i ) and outer ( o ) sites ., This configuration is denoted as Qi-Qi-bH-bL-c1-FeS-Qo-Qo ., Binding/dissociation of quinones in this model is accounted for by the replacement of reduced bound molecules at Qi by oxidized bound molecules and oxidized bound molecules at Qo by reduced bound molecules ( see Methods ) ., Numerical continuation of steady state solutions , as a function of succinate concentration , ( proportional to VmSDH , eq ., ( 1 ) ) , performed with CL_MATCONTL ( as described in Methods ) , revealed an interval of parameter values with multiple steady state solutions ( Fig . 3 , orange curves ) enclosed between two limit points ( LP ) indicating the points of fold bifurcations ., This interval of the parameter values for which multiple steady state solutions exist , corresponds to two curve segments of stable steady states and one curve segment of unstable steady states in between ., The method of numerical bifurcation analysis that we used allowed us to accurately and rigorously determine the bifurcation behavior of the whole system , without any simplifications ., At the same time , the main process underlying the bifurcation behavior can , in some cases , be heuristically identified by reducing a system by taking into account different time scales ., Such a reduction of Model 145 ( see Text S2 ) points to binding/dissociation of ubiquinone coupled with its reduction/oxidation as the main process responsible for the multistationarity of complex III ( Fig . S1 in Text S2 ) ., A gradual increase of succinate from low concentrations towards the interval of multistationarity leads to the lowest branch of steady states for quinol ( QH2 ) as shown in Fig . 3A ., Under such conditions complex III functions to give high electron flow and high ÄØ ( upper branches in Figs . 3B and 3C ) ., It should be noted that , although the concentrations of QH2 that constitute the lower branch of steady states in Fig . 3A are low , they nevertheless are sufficient to maintain high levels of ΔΨ ., The inset in Fig . 3C shows that ΔΨ drops to 0 as the concentration of succinate decreases to 0 ., This drop is a consequence of QH2 deficiency ., A decrease of succinate concentration from 15 to 0 corresponds to the almost linear decrease of QH2 levels from 0 . 02 to 0 nmol/mg prot ., The above described “active” state , that provides highest electron flow , is characterized by low levels of semiquinone ( SQ ) at the quinol oxidase site ( Qo ) ( Fig . 3D ) ., If succinate concentration surpasses the right limit point on Fig . 3A , the rate of ubiquinone reduction by succinate dehydrogenase outstrips the maximal capacity of its oxidation by complex III , therefore Q is almost completely converted into QH2 , as Fig . 3A shows ., This is the biochemical mechanism of the bifurcation ., The lack of an electron acceptor at Qi site results in a decrease of electron flow through the Q-cycle , a decrease of ΔΨ , and an increase in levels of SQ at Qo ., If then succinate concentration decreases , the blocked electron transport cannot produce a sufficient amount of acceptors Q to activate electron flow ., When the system is in such a blocked state , even decreased succinate dehydrogenase activity is sufficient to maintain low levels of Q and keep the system blocked ., If initially the electron carriers are oxidized , the solution of an initial value problem for the ODE system approaches an “active” steady state , and if initially the carriers are reduced , the solution approaches a “blocked” steady state ., Model 145 has 13 parameters ., The number of parameters is much less than the number of equations because parameters characterize the types of electron transport reaction , which is a much smaller number than the number of combinations of redox states of carriers ( the number of equations ) ., Different states participate in reactions of the same type with the same parameters , but they cannot be combined ( the system cannot be reduced without additional simplifications ) because the whole set of reactions for each state ( variable ) is different ., Fig . 3 shows curve segments of multiple steady state solutions corresponding to an interval of values of one parameter ., The shape and size of these curve segments depends on the values of other parameters ., The width of this interval may be smaller , or the interval may even disappear ., Fig . 3 shows an interval of relative succinate concentrations corresponding to multiple steady states , obtained using an algorithm ( described in Methods ) that scans all the parameters with the objective of finding as large interval as possible ., Including in our model all of the four configurations of complex III and an explicit description of quinone binding/dissociation increases the number of equations to 257 ( see Methods ) ., This more detailed model also has a region of multiple steady state solutions ., Application of the same algorithm maximizing the region of relative succinate concentration characterized by multiple steady states resulted in the interval that is larger than the one in the case of model 145 , as is shown for ΔΨ in Fig . 4 ., However , the qualitative behavior of the two models remains similar ., Evidently , model 145 faithfully accounts for the main properties of complex III determined by the Q-cycle mechanism ., Model 267 is obtained by adding to model 257 equations that account for reactions taking place in complex I ( described in Methods ) ., This extended model contains almost all of the essential components of the respiratory chain model that we used for the analysis of experimental data 12 ., Using it enables us to start the numerical bifurcation analysis for a “real” set of parameter values ( Table 2 ) that reproduces the measured dynamics of NADH reduction in the presence and absence of rotenone , and the maximal and state 4 respiration rates , when mitochondria are fueled by succinate or pyruvate/malate 12 ., Numerical continuation of steady state solutions for ΔΨ as a function of succinate concentration uncovers the existence of an interval of parameter values with multiple solutions , in the form of an S-shaped curve , enclosed between two limit points ( Fig . 5 ) ., There is also a Hopf bifurcation point in the vicinity of one limit point ., The sustained oscillation , which can be simulated in the parameter interval between the left limit point and Hopf bifurcation , has very small amplitude ( inset in Fig . 5 ) ., The mechanism of the fold bifurcations in this case is similar to that discussed for the simplest model ., Similar to the case presented in Fig . 3 , the stable steady states with the lowest values of ÄØ have the highest levels of SQ radicals at the Qo site; this may be a physical basis for high ROS production rates ., The measurements that were used to find the given set of parameters were performed in a suspension of isolated mitochondria ., The rate of electron transport from cytochrome c1 to cytochrome c is the parameter most affected by the procedure of isolation , since it depends on the structure of intermembrane space , which is significantly changed after the isolation ., In intact mitochondria the value of this parameter is expected to be higher than in isolated mitochondria ., Increasing its value by less than an order of magnitude increases the interval of multiple steady state solutions to infinity ( Fig . 6 ) ., Starting from an initially oxidized state , the system approaches a steady state , which , with numerical continuation with the substrate concentration as a parameter , results in the upper branch in Fig . 6 ., This branch does not contain any bifurcation points ., However , at a high substrate concentration , and starting from a reduced state , the system approaches another steady state located on a different branch marked blue in Fig . 6 ., The lower segment of this branch is stable ., A decrease of the substrate supply parameter ultimately leads to a limit point , and the upper segment of unstable steady states starts at this point ., Note , this branch of unstable steady states is not connected to the upper branch of stable steady states ., Similar to the cases analyzed above , the steady states corresponding to low ΔΨ values ( lowest blue branch ) are characterized by practically complete reduction of the free ubiquinone pool and maximal levels of free radicals SQ at the Qo site ., Similarly to the cases analyzed above , the steady states corresponding to high ΔΨ ( yellow branch ) are accompanied by an oxidized free ubiquinone pool and low levels of SQ at the Qo site ., The change of a single parameter that characterizes interaction between cytochromes c1 and c ( kc1c ) gives a qualitatively different behavior of model 267 , as seen in Fig . 6 from the comparison between the blue curve and the orange curve , which is redrawn from Fig . 5 ., Such a difference in behavior can be induced by swelling/shrinking of mitochondria , as was mentioned above , but also hypoxia/reoxygenation can induce a similar change of parameters and , thus , similar effects ., Hypoxia in our model can be simulated as a kc1c decrease ., Assume that before hypoxia the system effectively functions at some point on the yellow curve ( Fig . 6 ) ., Suppose the change of kc1c induced by hypoxia transforms the properties of the system so that the orange curve becomes the continuum of its steady states ., If before hypoxia the functional steady state was to the right of the rightmost limit point in orange curve , then after hypoxia the system evolves until it reaches a steady state in the lower segment of orange curve ( coinciding with the lower segment of blue curve ) ., If then the system is re-oxygenated , and the blue and yellow curves again become the continuum of steady states , it remains in the same state now located in the low segment of blue curve ., Thus , hypoxia and re-oxygenation change the state of the system ., Before hypoxia it generated high ΔΨ ( yellow curve in Fig . 6 ) and slowly produces ROS , whereas after re-oxygenation it stays in a state characterized by low ΔΨ ( blue curve with points in Fig . 6 ) and rapidly produces ROS ., Further extension of the ODE model to 272 equations , as described in Methods , by including the reactions of the TCA cycle with the parameters listed in Table 3 , allowed us to study the interaction of the respiratory chain with central carbohydrate metabolism ., In the extension , pyruvate is accounted for as a substrate for the TCA cycle , which provides succinate for complex II and NADH for complex I . Using parameter values verified by fitting the measured dynamics of NADH and respiration rates 12 , this model predicts the existence of wide range of pyruvate concentrations with two stable steady state segments ( Fig . 7 ) , as well as those described above with respect to succinate ., Similarly to the cases considered above , the redox state of free ubiquinone pool determines the dynamics of the system ., If the free ubiquinone pool ( Fig . 7A ) is not completely reduced , the electron flow through the respiratory chain ( Fig . 7B ) and ÄØ ( Fig . 7C ) is high , and the level of SQ at the Qo site ( Fig . 7D ) , determining the ROS production rate , is low ., On the other hand , if ubiquinone is practically completely reduced , the electron flow through the respiratory chain and ÄØ is low , and the level of SQ at the Qo site is high ., However , the bifurcations which determine a switch between the two steady state branches are different from those considered above ., Specifically , when the system is in an oxidized state , the increase of the pyruvate concentration leads to a Hopf bifurcation ., Oscillations in a neighborhood of this bifurcation point have insignificantly small amplitude ( see Fig . 8A ) ., An increase of the control parameter makes the amplitude greater ( inset in Fig . 8B ) , but the trajectory comes to the zone of attraction of the lower segment of stable steady states ( Fig . 7C ) and approaches one of these states ( Fig . 8B ) ., As the pyruvate concentration decreases , the system stays in this “reduced” stable curve segment until it reaches the limit point at ∼0 . 003 mM , where a curve segment of unstable steady states starts ., An increase of the cytochrome c1 to cytochrome c electron transition rate ( kc1c ) to 782 s−1 and an increase in VmSDH ( equation ( 5 ) ) from 171 ( Table 3 ) to 1714 nmol/s/mg changes the bifurcation behavior as is shown in Fig . 9A ., Although the bifurcation diagram here is also a basic S-shaped curve producing multi-stationarity , the entire segment of steady states between the two Hopf bifurcation points is unstable ., Stable oscillations of high amplitude appeared between these Hopf bifurcation points ( Fig . 9B and 9C ) ., ÄØ oscillates between 160 and 20 mV; such changes can be measured and , probably , this mechanism underlies the observed behavior 19 ., As Figs ., 9B and 9C show , ÄØ and the level of SQ at the Qo site ( defining the rate of ROS production ) oscillate in counter-phase ., This also corresponds to the behavior monitored in intact mitochondria 14 ., Variation of parameters can significantly change the durations of phases of low or high potential ( high or low ROS production rate , respectively ) ., This could be a basis of the ROS signaling 20 ., The mechanism of oscillations arises from an interaction of the ubiquinone reduction/oxidation with the TCA cycle ., The switch from the “oxidized” curve segment of steady states to the “reduced” one is accompanied by a decrease of the electron flow , and , as a consequence , an increase of the NADH levels ( decrease of NAD+ ) ., Since the conversion of pyruvate in the TCA cycle requires NAD+ , the production of succinate slows down ., The high levels of NADH are maintained for some time due to reverse electron transport through complex I reducing NAD+ ., A decrease of the substrate production in the TCA cycle and reverse electron transport result in an accumulation of a sufficient amount of the electron acceptor ubiquinone that activates electron transport , which results in switching back to the curve segment of the “oxidized” steady states , and then the cycle repeats again ., The change from multi-stationarity shown in Fig . 7 to an oscillatory behavior shown in Fig . 9 is , in part , the result of a change of the rate constant kc1c , which accounts for interaction between cytochromes c1 and c ., This rate constant depends on the volume of the intermembrane space , where the interaction takes place ., The intermembrane volume can be controlled experimentally in a suspension of isolated mitochondria , and the correspondence of the model predictions and the measured ROS production rates under variations of the intermembrane space can be experimentally verified ., The change of the ROS production rate ( characterized by the level of SQ at the Qo site ) , predicted for an “oxidized” state of model 272 with an increase of succinate concentration , is shown in Fig . 10A ., The shape of the lower curve segment of steady state concentrations of SQ bound at the Qo site depends on parameter kc1c ., Fig . 10A shows a superposition of curve segments of stable steady states obtained at two different values of kc1c ., At a low value of this parameter ( ∼260 s−1 as shown in Table 1 ) , increasing the succinate concentration above 1 mM takes the system past the Hopf bifurcation point ( similar to that shown in Fig . 7D ) , and it switches to the upper curve segment of SQ stable steady state concentrations ( blue curve ) ., Increasing kc1c shifts this Hopf bifurcation point to infinity , so that the upper branch of stable steady states , although it still exists , becomes inaccessible from the lower branch in the space of succinate concentrations ( orange curve ) ., In Fig . 10A , the lower branch of steady states obtained at a higher value of kc1c crosses the lower branch obtained at a lower value of kc1c ., At low succinate concentrations , the levels of SQ at the Qo site are higher when kc1c is high ., At high succinate concentrations , this relationship is reversed ., We have confirmed this experimentally , registering the rate of the ROS production as a measure of the SQ concentration at the Qo site ., It is expected that kc1c decreases if the intermembrane space increases , thus diluting the concentration of cytochrome c that is included implicitly in this parameter ., The light scattering technique allows measuring changes in the volume of the mitochondrial matrix and , implicitly , the intermembrane space ., Light scattering is higher in KCl than in sucrose of the same osmolarity ( Fig . 10B ) ., This indicates that the mitochondrial matrix is more compact in KCl than in a sucrose solution ., The outer membrane is permeable for both solutes , hence the total mitochondrial volume , which it restricts , must be the same ., Therefore the intermembrane space , estimated as the difference between total and matrix volumes , is greater in KCl media ., Thus , mitochondria incubated in KCl are characterized by lower values of kc1c than those incubated in sucrose ., The experimental results shown in Fig . 10C are consistent with the model ., Indeed , in the media with sucrose , the rates of ROS production driven by low succinate concentrations ( 100–500 uM ) are higher than those in KCl supplemented media ., In the range of succinate concentrations ≥500 uM , the situation was reversed: ROS production in KCl-based media exceeded that observed in sucrose-based media ., Bifurcation behavior , as revealed by the numerical bifurcation analysis of complex III models , is inherent in the Q-cycle mechanism of electron transport ., The main process underlying fold bifurcations in the considered models of complex III was found to be reduction/oxidation and coupled binding/dissociation of ubiquinone in accordance with the Q-cycle mechanism ( Text S2 ) ., We show here that a decrease of ÄØ accompanied by an increase of ROS production rate can take place as a consequence of perturbations in the respiratory chain operation ., The most critical element in such bifurcation behavior is that the same metabolite is reduced at the Qi site and oxidized at the Qo site ., The interaction of complex III with complex I increases the width of the maximal interval of multiple solutions , see Fig . 5 , or may even make it infinite , as shown in Fig ., 6 . The width of this interval is sensitive to the parameter ( kc1c ) that characterizes the combined processes of electron transport from cytochrome c1 to c and further , ultimately reducing molecular oxygen ., Thus , it can represent the availability of oxygen and , in this case , the comparison of Figs ., 5 and 6 , given in Results , demonstrates how hypoxia and reoxygenation may perturb the system to a state of a very high ROS production ., Further extending the model by including into it the reactions of the TCA cycle preserves the interval of parameters where multiple steady state solutions exist ., In particular , there are two stable steady states at the parameter values defined in 12 by fitting measured experimental data , as shown in Fig ., 7 . In experiments performed previously 10 , we confirmed that isolated mitochondria incubated with high succinate concentration can persist in one of two different steady states ., Mitochondria can be switched from a high ROS production state to a low one by transient incubation with ADP , and then back to a high ROS production state by transient hypoxia ., Another experimental confirmation of the predicted behavior of the electron transport chain is the consistency between the predicted curves of steady state levels of SQ at Qo attained at various concentrations of succinate for two different swelling conditions and the measured curves of ROS production rate ( Fig . 10 ) ., Stable oscillations that can be obtained at the parameter values in a neighborhood of the Hopf bifurcation point have insignificantly small amplitude , and the region of their stability appears to be so small that it is practically undetectable ., However , an increase of the values of the two parameters , which characterize succinate dehydrogenase activity and the rate of combined reactions upstream from cytochrome c1 , results in the appearance of an interval of succinate concentrations where high-amplitude oscillations exist and are stable ( Fig . 9 ) ., Feedback interaction of the multistationary respiratory chain operation with the TCA cycle creates oscillations ., NADH , as a common metabolite , provides such feedback ( see “Mechanism of oscillations” in Result section ) ., The parameters shown in Tables 1–3 that were used for model 272 were determined from the best fit to the data from experiments performed in vitro in isolated mitochondria ., One can expect that the volume of the intermembrane space increased after the procedure of isolation ., Such a change of the intermembrane space dilutes cytochrome c , and thus decreases the rate of interaction of cytochromes c1 and c ., Natural spatial variability of succinate dehydrogenase activity contributes to an uncertainty in its estimated value ., Our results show that the change in these parameters , which can be expected in mitochondria of living cells , compared to the isolated ones , results in an oscillatory mode of operation ., In fact , in mitochondria of living cells , flashes and oscillations of ROS production accompanied by the counter-phase changes of ΔΨ can be measured either as a response to laser excitation 13 , 14 , or as a spontaneous mitochondrial activity 15 , 19–25 ., Usually , the measured in vivo decrease of ÄØ that accompanies the ROS flashes was ascribed to either a ROS-induced mitochondrial permeability transition ( MPT ) 13 , 14 or a ROS-activated inner membrane anion channel 26 ., Our study opens a new direction in the investigation of the MPT that is of great importance for understanding intracellular signaling and regulation ., In particular , it can help to solve the question: why is respiration inhibited during the MPT , when ÄØ is low and cytochrome c still remains in the intermembrane space ?, Our hypothesis is that the MPT is secondary with respect to the change in the steady state of respiration; it happens when the electron transport chain switches to the “reduced” steady state , where respiration is inhibited by the mechanism considered above ., There is more evidence supporting this hypothesis ., Matrix pH is well known to be important for the MPT and models considering it as a main factor governing opening/closure of the MPT pore describe the observed events of the MPT 27 ., However , the link between the change of matrix pH and the MPT was described phenomenologically; the concrete mechanism of the pH effect on the MPT remains elusive ., The models presented here points to the mechanisms by which pH can affect electron transport: protons are explicitly involved in reduction/oxidation of ubiquinone , which is the main process defining the bifurcation ., Alkalinization of the matrix must slow down SQ reduction at Qi site and , thus , block electron transport and facilitate the switch to the “reduced” state ., If the change to a reduced steady state of the electron transport chain induces the MPT , this provides a concrete mechanism of pH-induced MPT , though it requires further investigation ., Moreover , in some cases , ROS sparks and a decrease of ΔΨ may be a direct consequence of the functional organization of the electron transport , and may not require the involvement of other mechanisms ., Thus , many experimentally observed effects , such as excessive ROS production after hypoxia/reoxygenation , or oscillations of ROS production and of ΔΨ , can be explained as a consequence of intrinsic properties of the respiratory chain and its interaction with the central metabolic pathways ., These qualitatively different modes of behavior are manifestations of the same mechanism of electron transport , determined by its quantitative characteristics ., Understanding the qualitatively different types of behavior requires a quantitative analysis of electron transport and the linked reactions of the central metabolism ., The method presented here can be used for such an objective ., However , the simplifications of reality used in our models should be taken into account ., In particular , they represent complex III as a monomer , whereas it is known that the functional form is dimeric in living cells 28 ., The functional link of monomers at the level of cyt bL was analyzed based on the edge-to-edge distance between cyt bL hemes of the two monomers 29 ., It was found that the intermonomer interaction can affect the rate of electron transport , especially in the energized states and when the bH heme is reduced because of a lack of electron acceptors at the Qi site ., Using our method to model the dimer would require solving ODE systems containing roughly the square of the number of equations that we analyze here ., Such systems can be constructed , but solving them would create computational problems ., Performing a preliminary analysis of a simplified model of the bc1 dimer containing cyt bL and bH , and Qo binding sites , we found that , despite intermonomer interactions , which quantitatively affect the kinetic behavior of complex III , qualitatively , the dimer demonstrates the same types of bifurcation behavior as the monomer in the situations analyzed in 29 ., Another limitation of our models concerns the values for the parameters ., The basic set of parameters shown in Tables 1–3 originally was taken from 30 and was then modified by fitting experimental data 12 ., In principle , the rate constants

Introduction, Results, Discussion, Methods

The mitochondrial electron transport chain transforms energy satisfying cellular demand and generates reactive oxygen species ( ROS ) that act as metabolic signals or destructive factors ., Therefore , knowledge of the possible modes and bifurcations of electron transport that affect ROS signaling provides insight into the interrelationship of mitochondrial respiration with cellular metabolism ., Here , a bifurcation analysis of a sequence of the electron transport chain models of increasing complexity was used to analyze the contribution of individual components to the modes of respiratory chain behavior ., Our algorithm constructed models as large systems of ordinary differential equations describing the time evolution of the distribution of redox states of the respiratory complexes ., The most complete model of the respiratory chain and linked metabolic reactions predicted that condensed mitochondria produce more ROS at low succinate concentration and less ROS at high succinate levels than swelled mitochondria ., This prediction was validated by measuring ROS production under various swelling conditions ., A numerical bifurcation analysis revealed qualitatively different types of multistationary behavior and sustained oscillations in the parameter space near a region that was previously found to describe the behavior of isolated mitochondria ., The oscillations in transmembrane potential and ROS generation , observed in living cells were reproduced in the model that includes interaction of respiratory complexes with the reactions of TCA cycle ., Whereas multistationarity is an internal characteristic of the respiratory chain , the functional link of respiration with central metabolism creates oscillations , which can be understood as a means of auto-regulation of cell metabolism .

The mitochondrial respiratory chain shows a variety of modes of behavior ., In living cells , flashes of ROS production and oscillations accompanied by a decrease of transmembrane potential can be registered ., The mechanisms of such complex behavior are difficult to rationalize without a mathematical formalization of mitochondrial respiration ., Our most complete model of mitochondrial respiration accounts for the details of electron transport , reproducing the observed types of behavior , which includes the existence of multiple steady states and periodic oscillations ., This most detailed model contains hundreds of differential equations , and such complexity makes it difficult to grasp the main determinants of its behavior ., Therefore the full model was reduced to a simplified description of complex III only , and numerical bifurcation analysis was used to study its behavior ., Then the evolution of its behavior was traced in a sequence of models with increasing complexity leading back to the full model ., This analysis revealed the mechanism of switching between the modes of behavior and the conditions for persistence in a given state , which defines ATP production , ROS signaling and destructive effects ., This is important for understanding the biochemical basics of many systemic diseases .

macromolecular assemblies, mathematics, bioenergetics, biophysics simulations, biology, energy-producing processes, differential equations, biophysics, systems biology, biochemistry, biophysic al simulations, calculus, computational biology, genetics and genomics

journal.ppat.1005254

Interferon-α Subtypes in an Ex Vivo Model of Acute HIV-1 Infection: Expression, Potency and Effector Mechanisms

The type I interferons ( IFNs ) are critical players in the innate immune response against viral infections ., Shortly after infection , these cytokines are rapidly induced , stimulating an antiviral state through the induction of hundreds of interferon-stimulated genes ( ISGs ) 1 ., This family of cytokines include IFNα , the first cytokine produced through recombinant DNA technology and tested in clinical trials against many infectious diseases 2 ., Notably , IFNα is a collective term for 12 unique IFNα proteins or subtypes expressed by 13 IFNA genes that are tandemly arrayed on human chromosome 9 ., However , most clinical trials only utilize recombinant IFNα2 , the subtype that is currently licensed for the treatment of hepatitis B virus ( HBV ) and HCV infection ., IFNα2 was also evaluated for reducing HIV-1 plasma viral loads during chronic infection ., However , the variable levels of efficacy observed 3–6 and the advent of potent and safer antiretroviral drugs reduced enthusiasm for the use of IFNα in the clinical management HIV-1 infection ., Two major developments in recent years renewed interest in IFNα as a therapeutic for HIV-1 infection: ( 1 ) the discovery of antiretroviral restriction factors , most of which are induced by IFNα 7; and ( 2 ) the improved prospects in achieving functional HIV-1 cure , which may be advanced through IFNα-based therapies 8 , 9 ., However , this renewed interest also raised unanswered questions on the basic biology of IFNα , including the biological consequences of having an expanded IFNA gene family 10 , 11 ., In fact , the relative expression , antiviral potency and restriction factor mechanisms employed by the various IFNα subtypes against HIV-1 infection remains unclear ., One potential advantage for the expansion of the IFNA gene family could be the diversification of regulatory elements , which would allow the infected host to differentially express IFNA genes in response to diverse stimuli ., Plasmacytoid dendritic cells ( pDCs ) are the primary producers of IFNα in vivo 12 , and exposure of pDCs to HIV-1 or HIV-1 infected cells resulted in a dramatic rise in IFNα production 13 , 14 ., Measurements of total IFNα proteins rely on antibodies that may have different binding affinities to the IFNα subtypes ., Furthermore , antibodies that can distinguish the various IFNα subtypes are not yet available ., IFNα expression is primarily regulated at the mRNA level 15 ., Innate sensing of viruses , for example through Toll-like receptors ( TLRs ) , results in a signaling cascade that leads to the activation and recruitment of transcription factors to the IFNA promoter ( s ) 16 ., Thus , quantitative real-time PCR ( qPCR ) is a standard procedure used by many laboratories to measure IFNA gene expression , with increasing recognition on the importance of obtaining IFNA subtype distribution for understanding retroviral pathogenesis 17 ., However , quantifying the expression of the different IFNA subtype genes is complicated by their high sequence homology ( 78 to 99% ) ., Nevertheless , IFNA subtype expression profiles of pDCs were evaluated using quantitative real-time PCR assays developed for each IFNA gene 15 , 18–20 ., Humanized mice exposed to TLR7 agonists showed prominent expression of IFNA2 and IFNA14 in pDCs 18 but other studies showed equal expression of all IFNA subtypes following TLR ligand stimulation 15 , 19 ., These discrepancies suggested that measuring IFNA distribution by qPCR may be difficult to reproduce across laboratories ., Moreover , performing 12 qPCR reactions for each IFNA subtype would not be ideal for limited biological samples ., The lack of a robust method to quantify IFNA distribution is therefore a significant hurdle in understanding the role of IFNA subtypes in human health and disease ., Functional diversification may be another evolutionary advantage for an expanded IFNα gene family ., Although all IFNα subtypes signal through the same type I interferon receptor ( IFNAR ) , the IFNα subtypes exhibited different binding affinities for the IFNAR-1 and IFNAR-2 subunits 21 , 22 ., This might result in different signaling pathways induced by IFNα subtypes 23 and in distinct expression patterns of ISGs in vitro 24 ., In vivo , mouse IFNα subtypes exhibited different potencies against herpes simplex virus 1 , murine cytomegalovirus , vesicular stomatitis virus ( VSV ) , influenza virus and Friend retrovirus 11 , 25 ., Altogether , the data indicate that the IFNα subtypes are not functionally redundant , raising the immediate question of which IFNα subtypes are most potent against HIV-1 ., An early study revealed that IFNα2 may be the most potent , but only 6 IFNα subtypes were evaluated against an X4-tropic , lab-adapted HIV-1 strain in the MT-2 T cell line 26 , thereby raising issues regarding physiological relevance ., IFNα is induced very early during HIV-1 infection 27 , and blocking IFNAR signaling in the SIV/rhesus macaque model resulted in higher viral loads and pathogenesis 28 ., The impact of the early IFNα response against HIV-1 most likely manifests in the gut-associated lymphoid tissue ( GALT ) , as it is the major site of early HIV-1 amplification and spread that leads to a massive depletion of CD4+ T cells 29 , 30 ., Prior success in infecting gut lamina propria mononuclear cells ( LPMCs ) with HIV-1 31 led to the development of the Lamina Propria Aggregate Culture ( LPAC ) model 32 , 33 ., The LPAC model allows for the robust infection of primary gut CD4+ T cells with CCR5-tropic HIV-1 strains , subsequently leading to CD4+ T cell depletion ., Importantly , this model allowed for HIV-1 infection studies without the confounding effects of non-physiologic T cell activation , as HIV-1 can efficiently infect gut CD4+ T cells without exogenous mitogens 29–31 ., Thus , the LPAC model is an ideal ex vivo platform to evaluate the relative potency of the various IFNα subtypes against HIV-1 ., Identifying the key effectors behind the anti-HIV-1 activity of IFNα could pave the way for the design of novel IFNα-based therapeutics ., The APOBEC3 proteins ( A3G , A3F , A3D and A3H ) , Tetherin/BST-2 and Mx2 were considered as bona fide HIV-1 restriction factors 7 , 34–37 ., These factors were proposed as effectors of the IFNα treatment effect based on correlative studies using IFNα clinical trial data 38 , 39 as well as cell culture data 35 , 40–42 ., However , their regulation by diverse IFNα subtypes in mucosal CD4+ T cells has not yet been explored ., APOBEC3 and Tetherin are counteracted by the HIV-1 Vif and Vpu , respectively 7 , but it is important to note that these interactions are saturable ., Induction of APOBEC3 and Tetherin expression may undermine the antagonism due to Vif and Vpu by offsetting the balance of these respective interactions ., Tetherin and Mx2 inhibit HIV-1 in the infected cell , leading to a reduction in virus release 34–37 ., In contrast , the APOBEC3 proteins are packaged into budding HIV-1 particles and inhibit replication in the next target cell by impeding reverse transcription and hypermutating reverse transcripts 43 , 44 ., Thus , a strong case for APOBEC3 activity could be made if reduced HIV-1 virion infectivity and increased G→A hypermutation were both detected ., We previously showed that treatment of Friend retrovirus-infected wild-type mice with IFNα reduced viral loads , but not in Apobec3 knock-out ( KO ) mice 45 ., Given the longstanding evolutionary conflict between mammalian hosts and retroviruses 46 , we hypothesized that the human APOBEC3 proteins may also act as effectors of IFNα treatment against HIV-1 in mucosal CD4+ T cells ., Here , we modeled the role of the IFNα subtypes during acute HIV-1 infection ., Using a novel next-generation sequencing-based method , we quantified the relative expression of the IFNα subtypes following HIV-1 exposure in pDCs , and determined the relative antiviral potency of each IFNα subtype in the LPAC model ., Moreover , we determined the induction profiles of known HIV-1 restriction factors following treatment with individual IFNα subtypes , and provide evidence that the APOBEC3 proteins may serve as key effectors for the antiviral activity of IFNα against HIV-1 ., Plasmacytoid DCs ( pDCs ) are the primary sources of IFNα in vivo , migrating to the GALT from the periphery during acute SIV infection 47 and accumulating in mucosal tissues during chronic HIV-1 infection stages 48 , 49 ., To date , the IFNα subtypes produced by pDCs following HIV-1 sensing remain unknown ., To determine the expression levels of each IFNα subtype , we designed 2 complementary assays using primers designed in the most conserved regions of the 13 IFNA genes ( S1 Fig ) ., Using these primers , total IFNA expression relative to the housekeeping gene GAPDH could be measured by qPCR , whereas IFNA subtype distribution could be quantified by next-generation sequencing ., We used negative selection to enrich pDCs from PBMCs from 4 healthy donors and exposed the cells to HIV-1 virions ( R5-tropic BaL strain ) for 6 hrs ( Fig 1A ) ., A 6 hr timepoint was chosen to ensure the viability of the pDCs , which significantly decline by 24 h post-culture 50 , while capturing the initial burst of IFNA expression following viral sensing ., Total IFNA expression was induced 485-fold in pDCs following HIV-1 exposure , but not in PBMCs lacking pDCs , confirming that pDCs are the main producers of IFNα ( Fig 1B ) ., We next quantified the relative abundance of each IFNα subtype in pDCs ± HIV-1 ., Primers in the conserved regions were modified with Illumina-sequencing adaptors , and the IFNA subtype designation for each sequence was determined based on the polymorphic regions in the amplicon ., IFNA1 and IFNA13 encode identical proteins and had identical DNA sequences in the region amplified , so these genes were counted together as IFNA1/13 ., On average , 9 , 543 IFNA sequence reads were analyzed per donor per condition ., The IFNA genes were aligned according to their relative genomic positions and their proportional expression values are shown ( Fig 1C ) ., The proportional expression of different IFNA subtypes by pDCs from different donors was very consistent both in naïve cultures ( Fig 1D ) and following HIV-1 exposure ( Fig 1E ) ., Interestingly , there was a strong bias towards expression of IFNA genes at the centromeric half of the IFNA complex following HIV-1 exposure ( Fig 1E ) ., Five out of six IFNA genes in this genomic cluster accounted for >70% of the IFNA subtypes expressed by pDCs following HIV-1 exposure ( Fig 1F ) ., The exception was IFNA6 , which decreased as a percentage of the total IFNA ., The augmented IFNA subtype expression levels were independent of genomic orientation , as IFNA2 and IFNA8 were both highly expressed yet had opposite genomic orientations ( Fig 1C ) ., We then determined the absolute copy numbers of each IFNA subtype by multiplying the percentage values ( Fig 1D and 1E ) with the total copy numbers ( Fig 1B ) ., The absolute copy numbers of all IFNA subtypes increased in pDCs following HIV-1 exposure , though to varying degrees ( S2 Fig ) ., IFNA14 , IFNA2 and IFNA10 were induced over 1000-fold following HIV-1 exposure of pDCs , whereas IFNA6 was induced by ~100-fold ., Overall , the results revealed a pattern of IFNA gene induction after HIV-1 exposure that appeared to be linked to chromosomal position ., Since the GALT is the major site of early HIV-1 amplification and spread , we utilized LPAC as a physiologically relevant model to determine the relative anti-HIV potency of each IFNα subtype ., In particular , we were interested in whether IFNα2 , the subtype approved for clinical use , was the optimal IFNα subtype for inhibiting HIV-1 ., Fig 2A outlines the experimental infection protocol ., Analyzing the HIV-1 potency of all 12 IFNα subtypes at multiple doses was not feasible in the LPAC model because of the limited number of LPMCs available per donor ., Thus , initial dose-response tests were performed with IFNα14 , which potently inhibited HIV-1 in a pilot experiment ., Following infection with HIV-1BaL , LPMCs were rinsed with culture media and resuspended to various IFNα14 concentrations ., Infection levels were evaluated at 4 days post-infection ( dpi ) to capture not only the impact of restriction factors that inhibit HIV-1 virus production , but also those that inhibit virion infectivity , which would decrease infection after one round of replication ( S3 Fig ) ., The percentage of infected CD4+ T cells was measured by detecting intracellular HIV-1 p24 capsid expression by flow cytometry , as we previously described 32 , 33 ., To account for HIV-1 Nef and Vpu-mediated CD4 downregulation 51 , we gated on CD3+CD8- cells ., A screen of LPMCs from 7 donors revealed that IFNα14 restricted productive HIV-1 infection , and that the inhibition was saturable at higher concentrations ( Fig 2B ) ., The majority of the LPMC donors had similar sensitivity to IFNα14-treatment , with the exception of one donor who responded to lower concentrations ., An IFNα concentration of 100 pg/ml was in the linear range of the dose response curve ( ~50% inhibition ) , and was chosen for the subsequent evaluation of all IFNα subtypes in 4 LPMC donors ., This concentration was also within the range of IFNα levels in plasma following HIV-1 infection in vivo 52 ., Majority of the cells in the LPMC donors used were CD3+ T cells ( 88% ± 3% ) ., On average , 65% of the LP T cells were CD4+ ., Myeloid DCs and gamma-delta T cells account for <1% of the total LPMC subpopulations , respectively ., Recombinant IFNα subtypes were added to LPMCs ( 100 pg/ml ) after spinoculation ( Fig 2A ) ., At 4 dpi , HIV-1 infected cells were quantified by detecting intracellular p24 by flow cytometry as in Fig 2B ., There were clear differences in the potency of the IFNα subtypes in inhibiting productive HIV-1 infection ( Fig 2C ) ., IFNα8 , IFNα14 and IFNα6 showed the highest levels of inhibition , whereas IFNα1 and IFNα2 had no significant effect ., The supernatants were also tested for infectious HIV-1 titers using the TZM . bl assay ( S3 Fig ) ., Again , the same 3 IFNα subtypes were most potent , whereas IFNα1 remained the least potent ( Fig 2D ) ., The antiviral potency of the different IFNα subtypes as measured by p24 flow cytometry and the TZM . bl assay significantly correlated with each other ( S4A Fig ) ., Although IFNα2 had no significant effect on cellular HIV-1 infection ( Fig 2C ) , it moderately reduced infectious titers ( Fig 2D ) ., Overall , the LPAC data revealed differences in the potencies of IFNα subtypes in inhibiting HIV-1 infection ., IFNα2 , the current subtype approved for clinical use , was one of the least potent subtypes ., To investigate whether the IFNα response of pDCs following HIV-1 exposure was biased towards the expression of the most potent antiviral IFNα subtypes , we next determined the relationship between IFNα subtype expression levels and relative potency ., Absolute IFNA subtype copy numbers were calculated by multiplying the total IFNA copies ( Fig 1B ) by the percentage of total IFNA for each subtype ( Fig 1E ) ., This provided an estimated copy number of each IFNA subtype per 106 copies of GAPDH ., Using these values , a significant inverse correlation was observed between IFNA subtype expression and potency ( Fig 3A and S4B Fig ) ., This correlation can be exemplified as follows ., IFNα1 was highly expressed but ineffective at inhibiting HIV-1 replication ., IFNα6 , one of the most potent subtypes , was among the least abundant following HIV-1 exposure ., IFNα2 showed a very high fold-increase following HIV-1 exposure relative to baseline but had weak antiviral efficacy ., IFNα5 is expressed at higher relative abundance ( Fig 1F ) but was also weakly antiviral ., These results revealed that the predominant IFNA subtypes produced by pDCs following HIV-1 exposure had low antiviral potency ., Two notable exceptions were IFNα8 and IFNα14 , which exhibited strong anti-HIV-1 activity and also had high expression in pDCs following HIV-1 exposure ( Fig 3A ) ., Exclusion of the IFNα8 and IFNα14 datapoints further strengthen the inverse correlation ( R2 = 0 . 62 , p = 0 . 007 ) ., Data from the Schreiber group 22 revealed that different IFNα subtypes exhibited variable binding affinities to IFNAR as estimated by surface plasmon resonance against each subunit , IFNAR-1 and IFNAR-2 ., We therefore determined if IFNα subtype anti-HIV-1 potency ( Fig 2C and 2D ) correlated with published binding affinity data to IFNAR 22 ., There was a significant positive correlation between antiviral potency and binding affinity ( KA ) to IFNAR-2 ( Fig 3B and S4C Fig ) , but not the IFNAR-1 subunit ( Fig 3C and S4D Fig ) ., These analyses suggested that following HIV-1 exposure , pDCs produced IFNα subtypes with relatively low antiviral activity and lower binding affinity to IFNAR-2 ., In particular , IFNα1 was expressed at high levels by pDCs exposed to HIV-1 virions but had the weakest IFNAR-2 binding affinity and the lowest anti-HIV-1 potency in the LPAC model ., The correlation between antiviral potency and IFNAR binding affinity suggested that the more potent IFNα subtypes might trigger higher ISG induction ., To test this hypothesis , we quantified the mRNA expression levels of the IFNα-inducible HIV-1 restriction factors Mx2 , Tetherin and APOBEC3 in LP CD4+ T cells after stimulation with representative IFNα subtypes ., We focused on CD4+ T cells , the major cellular targets of HIV-1 replication in the GALT , but not intestinal macrophages , which are non-permissive to HIV-1 infection 53 ., We selected IFNα8 and IFNα14 as potent IFNα subtypes due to their high affinity , highest antiviral potency in the LPAC model and high expression level in pDCs ., IFNα1 and IFNα2 were selected as weak IFNα subtypes due to their relatively low affinity , weaker antiviral activity in the LPAC model ( with IFNα2 being more potent than IFNα1 ) , but high expression level in HIV-1-exposed pDCs ( IFNα1 and IFNα2 ) ., IFNα2 was also chosen because of its clinical relevance ., LPMCs were infected with HIV-1BaL and 100 pg/ml IFNα was administered ., After 24 hr , CD4+ T cells were negatively selected and ISG mRNA expression was evaluated by qPCR ( Fig 4A ) ., The magnitude of ISG induction was donor-dependent so the data for each donor are presented ., ( Fig 4B to 4E ) ., The ISG expression that best correlated with the relative antiviral activities of the IFNα subtypes was Mx2 ( Fig 4B ) ., IFNα8 ( 3 of 3 donors ) and IFNα14 ( 2 of 3 donors ) more significantly induced Mx2 compared to IFNα1 and IFNα2 ., IFNα2 , which showed moderate antiviral activity ( Fig 2D ) , more significantly induced Mx2 compared to IFNα1 in 3 of 3 donors ., Tetherin induction exhibited trends similar to Mx2 , but the differences were not as consistent between donors ( Fig 4C and 4D ) ., Overall , the more antiviral IFNα subtypes induced Mx2 and Tetherin to higher levels ., In contrast , A3G ( Fig 4E ) was not significantly induced by any of the IFNα subtypes ., A3F and A3D expression were induced in a few cases with IFNα treatment ( S6 Fig ) , but the induction levels did not correlate with the relative anti-HIV-1 potency of the IFNα subtypes ., We previously demonstrated that mouse Apobec3 was the primary effector of IFNα treatment against Friend retrovirus infection despite not being transcriptionally induced 45 ., We therefore investigated the potential contribution of human APOBEC3 proteins to the IFNα-treatment effect ., The APOBEC3 proteins A3G , A3F , A3D and A3H do not inhibit HIV-1 in the producer cell ., Instead , these proteins get packaged into HIV-1 virions and inhibit replication in the next target cell ., Thus , non-infectious virion release is a distinguishing feature of APOBEC3-mediated retrovirus restriction 54 , 55 ., By contrast , most restriction factors such as Mx2 and tetherin inhibit virus particle production in the infected cell 7 ., Virion infectivity is typically measured by determining the ratio of infectious titer as measured by the TZM . bl assay and the total viral particles released in the supernatant as measured by HIV-1 p24 ELISA ( S3 Fig ) ., LPMCs from 6 donors were infected with HIV-1BaL and were treated with IFNα1 , IFNα2 , IFNα8 and IFNα14 ., At 4 dpi , all 4 IFNα subtypes inhibited virus particle release to the same extent ( Fig 5A ) ., By contrast , the infectious titers were reduced significantly more by IFNα8 and IFNα14 compared to IFNα1 and IFNα2 ( Fig 5B ) ., Thus , inhibition of virion infectivity correlated with the antiviral efficacy of the IFNα subtypes ( Fig 5C ) ., In particular , IFNα8 and IFNα14 were the most potent at inhibiting virion infectivity whereas IFNα1 had no significant effect ., In order to confirm that the findings were not specific to HIV-1BaL , LPMCs were infected with transmitted/founder ( T/F ) HIV-1 strains , which are infectious molecular clones reconstructed from acute HIV-1 infection samples 56–58 ., In 6 LPMC donors , the antiretroviral activity of IFNα1 and IFNα8 against the T/F HIV-1 strains CH470 , CH40 , and CH58 were compared ., IFNα1 and IFNα8 inhibited virus particle release to similar extents for CH40 and CH58 ( Fig 5D ) , whereas CH470 particle release was slightly more inhibited by IFNα8 ., In virion infectivity assays , IFNα8 more potently inhibited the 3 T/F HIV-1 strains ( Fig 5E ) ., We also evaluated the impact of IFNα8 in 13 additional T/F HIV-1 strains in 2 LPMC donors ., IFNα8 treatment resulted in a highly significant ( ~4-fold ) decrease in virion infectivity ( Fig 5F ) ., IFNα14 treatment also significantly inhibited the virion infectivity of these T/F HIV-1 strains ( S5 Fig ) ., These data indirectly suggested that the more potent IFNα subtypes augmented APOBEC3-mediated restriction of multiple HIV-1 strains ., The APOBEC3 proteins A3F , A3D and A3H mutated HIV-1 reverse transcripts with a preferred TC context , leading to GA→AA mutations in the retroviral plus strand , whereas A3G preferentially mutated in the CC context , leading to proviral GG→AG mutations 59 ., Thus , the magnitude of retroviral mutations in the GA→AA versus GG→AG context could be used to determine the APOBEC3 members responsible for HIV-1 G-to-A hypermutation and to provide additional evidence of APOBEC3 involvement in HIV-1 restriction ., To quantify APOBEC3-mediated retroviral mutations , we recently developed a next-generation sequencing approach to quantify mouse retrovirus hypermutation 60 ., To extend this method to HIV-1 , we designed barcoded Illumina primers encompassing gp41/nef ( 420–450 bp depending on the strain ) , a region that may be more susceptible to APOBEC3-mediated deamination due to longer retention in single-stranded form during reverse transcription 61 ., We initially tested the method by infecting LPMCs with WT HIV-1 NL4-3 and NL4-3ΔVif , which cannot counteract the effects of APOBEC3 ., The percentage of GG→AG and GA→AA mutations were computed against the mutations at C or G bases , which are directly modified by deaminases ., As expected , there was a significant increase in GG→AG and GA→AA mutations in NL4-3ΔVif compared to WT at 4 dpi ( Fig 6A ) ., Thus , A3G and A3F/A3D/A3H actively mutated HIV-1ΔVif in gut CD4+ T cells ., Following the validation of the next-generation sequencing method , we next analyzed proviral HIV-1 sequences for evidence of GG→GA and GA→AA mutations following treatment with IFNα8 or IFNα1 ., LPMCs were infected with T/F HIV-1 strains CH470 , CH40 , and CH58 ., These strains were derived from infectious molecular clones and therefore allow for straightforward mutational analysis ., These 3 HIV-1 strains also had reduced virion infectivity following IFNα8 but not IFNα1 treatment ( Fig 5E ) ., Untreated and IFNα-treated infected cells were harvested at 4 dpi ., Sequences were pooled for each of the HIV-1 CH470 , CH40 , and CH58 strains , respectively , to allow for a thorough analysis of mutational patterns ., A 2×2 contingency analysis was performed to test if IFNα had any effect on A3F/D/H-type ( GA→AA ) or A3G-type mutations ( GG→AG ) relative to the total number of C or G mutations ., Following IFNα8 treatment , both GG→AG and GA→AA mutations significantly increased in CH470 ( Fig 6B ) ., GG→AG mutations also significantly increased in CH40 , and to a lesser extent in CH58 ( Fig 6B ) ., Surprisingly , IFNα1 treatment also increased GG→AG mutations in CH40 , CH58 and CH470 ( Fig 6C ) ., Thus , both IFNα8 and IFNα1 treatment increased proviral DNA mutations that were associated with A3G deaminase activity ., Acute HIV-1 infection is characterized by extensive virus replication in the GALT , suggesting that the innate immune response could have a considerable impact on early HIV-1 spread in this compartment ., In particular , IFNα exhibited potent anti-HIV-1 properties in vitro and was one of the first cytokines induced during acute HIV-1 infection 27 ., Blocking type I IFN signaling in the SIV/rhesus macaque model resulted in more severe pathogenesis 28 ., T/F HIV-1 strains exhibited higher resistance to type I IFNs than counterpart chronic strains , suggesting that type I IFNs exerted a strong selective pressure during acute HIV-1 infection 57 , 58 ., These studies suggested that the initial IFNα response may serve as a roadblock for HIV-1 replication and spread in the GALT ., However , there were 12 IFNα subtypes , and to date , it remained unknown which IFNα subtypes were produced by pDCs , the professional IFNα-producing cells that rapidly migrate and reside in the GALT following HIV-1/SIV infection 47–49 ., Moreover , only one subtype , IFNα2 , was evaluated in clinical trials to reduce HIV-1 viremia ., In fact , the clinical use of IFNα2 was largely driven by its status as the first IFNα subtype cloned for large-scale production 2 , and not from a systematic evaluation of antiviral potencies in physiologically-relevant target cells ., Thus , the current study was undertaken to investigate the relative expression of the different IFNα subtypes in pDCs and their antiviral potency in the LPAC model ., A major finding from this work was that IFNα8 , IFNα6 and IFNα14 were the most effective at inhibiting HIV-1 replication in gut CD4+ T cells ., By contrast , the antiviral activity of IFNα2 was weak at best ., IFNα8 , IFNα6 , and IFNα14 exhibited strong binding affinities to IFNAR-2 22 ., This suggests that binding affinity to IFNAR-2 , proposed as the first IFNAR subunit that binds IFNα 62 , may contribute to the differential potencies of the IFNα subtypes ., This notion was corroborated by the higher ISG induction profile for IFNα8 and IFNα14 compared to IFNα2 ., Sequence analyses of IFNα8 , IFNα6 and IFNα14 in human populations revealed that DNA polymorphisms in these subtypes tend to preserve the amino acid sequence ( e . g . , purifying selection ) 63 , suggesting that these IFNα subtypes may have essential roles in vivo ., Moreover , IFNα8 exhibited strong antiviral activity against other viruses 64 ., Interestingly , using a novel method to quantify IFNA subtype distribution , we observed an inverse correlation between IFNα subtype expression in HIV-1-exposed pDCs and anti-HIV-1 potency ., IFNα6 fit this trend–it was one of the least expressed IFNα subtypes in HIV-1-exposed pDC cultures ., IFNα6 was also weakly expressed by pDCs stimulated with TLR ligands 15 ., However , IFNα8 and IFNα14 were both potent and more abundantly produced by pDCs exposed to HIV-1 ., IFNα8 and IFNα14 were encoded within the centromeric half of the IFNA complex , suggesting that epigenetic mechanisms may regulate their expression ., The data suggest that IFNα8 and IFNα14 may constitute the most potent antiviral fraction of the initial IFNα response against HIV-1 infection ., However , it should be noted that IFNα8 and IFNα14 only account for ~20% of the total IFNA transcripts produced by pDCs following HIV-1 exposure ., The majority of the IFNα subtypes expressed by pDCs following HIV-1 exposure had relatively weak antiviral activity ( IFNA1 , 2 and 5 account for >40% of IFNA transcripts ) ., In particular , the most expressed IFNα subtype , IFNα1 , had the weakest antiviral activity ., IFNα1 also exhibited very weak activity against VSV and HCV , and the lowest binding affinity for IFNAR-2 22 , 64–66 ., IFNα2 was also highly induced in pDCs post-HIV-1 exposure , consistent with another study showing IFNα2 was upregulated in HIV-1-infected individuals 67 ., We speculate that IFNα1 and IFNα2 induction may be a strategy used by HIV-1 to evade a more potent IFNα response ., However , the rationale for why humans evolved weakly antiviral IFNα subtypes in the first place remains unknown ., One possibility is that weakly antiviral IFNα subtypes may be better at modulating other immunological processes ., If true , then these IFNα subtypes could potentially elicit more adverse effects if administered therapeutically ., IFNα2 therapy was long known to have undesirable clinical side-effects including fever , fatigue and lymphopenia 2 ., Moreover , in an intriguing paradox , high IFNα expression levels during chronic HIV-1 infection correlated with disease progression 52 , 68 ., This led some to propose blocking IFNα signaling in chronic HIV-1-infected individuals to reduce immune activation 69 ., However , the IFNα subtypes responsible for the link between IFNα and chronic immune activation remains unknown ., The development of the IFNA subtyping method described here should facilitate revisiting this phenomenon ., In addition , further studies would be required to evaluate the tolerability profile of IFNα8 , IFNα6 and IFNα14 relative to IFNα2 and IFNα1 ., One possible strategy to harness the antiviral properties of IFNα for the design of safer HIV-1 therapeutics is to focus on its downstream antiviral effectors ., Many ISGs were reported to have inhibitory activity against HIV-1 in vitro 70 , but transcriptional induction levels may not predict the most potent antiviral effectors of IFNα 45 ., In this study , the more antiviral IFNα subtypes induced Mx2 and Tetherin to a greater extent ., Mx2 and Tetherin act on the producer cell , decreasing viral production ., Thus , if the IFNα subtypes were acting through these restriction factors to inhibit HIV-1 replication , we would expect higher inhibition of virus production by the more potent IFNα subtypes ., Surprisingly , this was not the case: IFNα1 inhibited virus particle production to a similar extent as IFNα8 and IFNα14 ., Thus , Mx2 or Tetherin may not be mediating the differences in antiviral potencies between the IFNα subtypes ., In other words , the differential induction of Mx2 and Tetherin expression by potent versus weak IFNα subtypes may just reflect the magnitude of IFNAR signaling and not necessarily indicate the mobilization of these effector mechanisms ., The IFNα subtypes did not significantly upregulate A3G , A3F and A3D transcription in gut CD4+ T cells , consistent with previous data using IFNα in PBMCs 71 , 72 ., Nonetheless , the relative potencies of the IFNα subtypes were associated with reduced virion infectivity , thus pointing to the APOBEC3 proteins as a significant antiviral effector of IFNα ., The notion that the APOBEC3 proteins could act as significant effectors of potent IFNα subtypes makes evolutionary sense based on our studies in mice 45 ., However , the mechanism for how IFNα improved APOBEC3 function without transcriptional induction remains to be determined ., Surprisingly , both the potent ( IFNα8 ) and weak ( IFNα1 ) subtypes induced retroviral GG→AG hypermutation , suggesting that the deaminase-dependent activity of A3G did not correlate with the relative antiretroviral potencies of the IFNα subtypes ., A3G inhibits HIV-1 through a deaminase-independent and deaminase-dependent mechanism ., The deaminase-independent mechanism acts upstream by inhibiting the elongation of reverse transcripts ,

Introduction, Results, Discussion, Materials and Methods

HIV-1 is transmitted primarily across mucosal surfaces and rapidly spreads within the intestinal mucosa during acute infection ., The type I interferons ( IFNs ) likely serve as a first line of defense , but the relative expression and antiviral properties of the 12 IFNα subtypes against HIV-1 infection of mucosal tissues remain unknown ., Here , we evaluated the expression of all IFNα subtypes in HIV-1-exposed plasmacytoid dendritic cells by next-generation sequencing ., We then determined the relative antiviral potency of each IFNα subtype ex vivo using the human intestinal Lamina Propria Aggregate Culture model ., IFNα subtype transcripts from the centromeric half of the IFNA gene complex were highly expressed in pDCs following HIV-1 exposure ., There was an inverse relationship between IFNA subtype expression and potency ., IFNα8 , IFNα6 and IFNα14 were the most potent in restricting HIV-1 infection ., IFNα2 , the clinically-approved subtype , and IFNα1 were both highly expressed but exhibited relatively weak antiviral activity ., The relative potencies correlated with binding affinity to the type I IFN receptor and the induction levels of HIV-1 restriction factors Mx2 and Tetherin/BST-2 but not APOBEC3G , F and D . However , despite the lack of APOBEC3 transcriptional induction , the higher relative potency of IFNα8 and IFNα14 correlated with stronger inhibition of virion infectivity , which is linked to deaminase-independent APOBEC3 restriction activity ., By contrast , both potent ( IFNα8 ) and weak ( IFNα1 ) subtypes significantly induced HIV-1 GG-to-AG hypermutation ., The results unravel non-redundant functions of the IFNα subtypes against HIV-1 infection , with strong implications for HIV-1 mucosal immunity , viral evolution and IFNα-based functional cure strategies .

The therapeutic potential of recombinant IFNα against HIV-1 infection has been explored for 25 years , but its effectiveness was inconsistent ., However , these clinical trials administered IFNα2 , which is only one member of a 12-protein family of IFNα subtypes ., More recently , IFNα was found to activate ‘restriction factors’–proteins that can directly inhibit HIV-1 ., To date , it remains unknown which IFNα subtypes are produced by professional IFNα producing cells known as plasmacytoid dendritic cells and which IFNα subtypes are more effective in inhibiting HIV-1 infection in the gastrointestinal tract , the primary site of early HIV-1 replication ., Here , we show that weaker IFNα subtypes were more highly expressed following HIV-1 infection ., Using an infection platform that captures important characteristics of early HIV-1 infection in the gut , several IFNα subtypes were found to be more effective at inhibiting HIV-1 than IFNα2 ., In particular , IFNα8 and IFNα14 more potently reduced the infectivity of HIV-1 virions , an activity that can be attributed to the APOBEC3 proteins ., Our findings strongly support the evaluation of potent IFNα subtypes in currently evolving HIV-1 curative strategies .

journal.pcbi.1000656

FluTE, a Publicly Available Stochastic Influenza Epidemic Simulation Model

Mathematical and computer models of epidemics have contributed to our understanding of the spread of infectious disease and the measures needed to contain or mitigate them 1–9 ., Detailed computer simulations will play an important role in evaluating containment and mitigation strategies for future epidemics 8 ., Although many simulation models have been described in the literature , few are publicly available ., Releasing the source code of models would allow others to evaluate the quality of the simulation , replicate results , and alter and improve the model ., We have released the source code for a new stochastic model of influenza epidemics , FluTE ., FluTE is an individual-based model capable of simulating the spread of influenza across major metropolitan areas or the continental United States ., The models structure is based on previously published work 3 , 6 , but FluTE incorporates a more sophisticated natural history of influenza , more realistic intervention strategies , and can run on a personal computer ., Here , we describe the new model and illustrate how it can be used to study the dynamics of an epidemic and to investigate the population-level effects of interventions ., The simulation creates synthetic populations based on typical American communities ., The population is divided into census tracts , and each tract is subdivided into communities of 500–3000 individuals based on earlier models 6 , 10 ., Each community is populated by randomly generated households of size 1–7 using the US-wide family size distribution from the 2000 Census ( Table 1 ) ., The household is the closest social mixing group , within which contacts between individuals occur most frequently and thus influenza is transmitted most often ., The population is organized as a hierarchy of increasingly large but less intimate mixing groups , from the household cluster ( sets of four socially close households ) , neighborhoods ( 1/4 of a community ) , and the community ., Although the model results are not sensitive to the exact size of these groups , including such groups creates a realistic contact network for disease transmission 11 ., At night , everyone can make contact with other individuals in their families , household clusters , home neighborhoods , and home communities ., In the daytime , individuals might interact with additional groups ., During the day , most children attend school or a playgroup , where there is a relatively high probability of transmission ., Preschool-age children usually belong to either a playgroup of four children or a neighborhood preschool , which typically has 14 students ., Each community has mixing groups that represent two elementary schools , one middle school , and one high school , which typically have 79 , 128 , and 155 students , respectively ., Most working-age adults ( about 72% of 19–64 year-olds ) are employed ., Employment rates are determined on a tract-by-tract basis using data from the US Census 2000s Summary File 3 , table PCT35 ., Employed individuals often work outside of their home communities ., Each employed individual is assigned to work in a destination census tract based on commuting data taken from Part 3 of the Census Transportation Planning Package ( http://www . fhwa . dot . gov/ctpp/dataprod . htm ) , which provides information on the home and destination census tracts of workers in the United States ., We eliminated commutes over 100 miles from the data as in 6 because many of these trips represent sporadic long-distance travel rather than daily commutes ., Working individuals are assigned to communities and neighborhoods within their destination tracts to simulate casual community contacts during the day , and a work group of about 20 people to represent their close contacts at the workplace ., Unemployed individuals remain in their home communities and do not have close daytime contacts except with members of their households who are not employed or enrolled in school ., Individuals can engage in short-term , long-distance domestic travel to represent vacations and other trips ., Travel in our model is based on the implementation in 6 , which uses data from the 1995 American Travel Survey data available from the U . S . Department of Transportation , Bureau of Transportation Statistics ( http://www . bts . gov/publications/national_transportation_statistics/ ) ., Each day , an individual has a fixed probability of starting a trip based on an age-specific probability of traveling: 0 . 0023 for 0–4 year olds , 0 . 0023 for 5–18 , 0 . 0050 for 19–29 , 0 . 0053 for 30–64 , and 0 . 0028 for 65 and older ., The traveler will stay at the destination for 0–11 nights , with 23 . 9% of trips lasting for a single day ( and no nights ) , 50 . 2% including 1–3 nights away , 18 . 5% including 4–7 nights away , and 7 . 4% for 8–11 nights ., We do not include differences in travel frequency or duration during different times of the year ( e . g . , summer and holiday trips ) ., The destination is a randomly selected census tract , in which a random community , neighborhood , and workplace ( if the traveler is between 19 and 64 years old ) are assigned to be the travelers mixing groups ., A random member of this community is assigned to be the travelers contact person , and at night the traveler will behave as if he/she belongs to the contacts household , household cluster , and neighborhood ., The traveler may withdraw to this household if ill ., The exact implementation of short-term , long-distance travel is not important , but some long-distance travel is required in large populations for the epidemic to spread in a realistic manner ., For simulations of smaller regions , such as a single county , there is no need to include long-distance travel ., New infected individuals are introduced to a simulation by infecting randomly selected people ., This epidemic seeding process can occur once at the beginning of a simulation or daily ., In addition , one can simulate an epidemic that is seeded from international travelers ., In this scenario , randomly selected individuals in the counties with one of the United States 15 busiest international airports are infected each day , proportional to the daily traffic of these airports ( see Table 2 ) ., The current modeling of the natural history of influenza is as follows: An individual is infectious for six days starting the day after becoming infected ., The individuals infectiousness is proportional to the log of the daily viral titers taken from a randomly chosen one of the six experimentally infected patients described in 12 , 13 ( Figure 1 ) ., An individual is asymptomatic during the incubation period , which lasts from one , two , or three days ( with 30% , 50% , and 20% probabilities , respectively ) ., After incubation , the individual has a 67% chance of becoming symptomatic 14 , 15 ., Symptomatic individuals are twice as infectious as asymptomatic people and may withdraw to the home after 0 to 2 days 16 ( with probabilities summarized in Table 3 ) ., People who withdraw interact only with their households ., Six days after infection , an individual recovers and is no longer susceptible ., The simulation runs in discrete time , with two time steps per simulated day to represent daytime and nighttime social interactions ., The contact probability of two individuals in the same mixing group is the probability that they will have sufficient contact for transmission during a time step ., Contact probabilities of individuals within families were tuned so that the simulated household secondary attack rates match estimates from 17 ( Table 4 ) ., Contact probabilities within other mixing groups were tuned so that the final age-specific illness attack rates were similar to past influenza pandemics ( Table 5 ) , particularly Asian A ( H2N2 ) and 2009 novel influenza A ( H1N1 ) influenza , and the percentage of transmissions that can be attributed to each mixing group matched those in 6 , 18–20 , although these values depend on the transmissibility ( ) of the disease ( Table 6 ) ., These contact probabilities are in general agreement with other simulation models 8 and with a recent study of physical contacts between individuals 21 ., Contact probabilities for all types of mixing groups are summarized in Table 7 ., Transmission probabilities in the simulation are adjusted by multiplying all contact probabilities by a scalar , , to obtain the desired , the basic reproductive number , which is defined as the average number of secondary infections from a typical infected individual in a fully susceptible population 22 ., To derive the relationship between and , we infected a single randomly selected person in an otherwise fully susceptible 2000-person community with a 74% working-age adult employment rate and counted the number of individuals that person infected , repeating this procedure 1 , 000 times for several values of ., The relationship between the average number of secondary cases was approximately linear for a biologically plausible range of values: ( Figure 2 ) ., However , the average number of secondary cases was higher when the index case was a child because children tend to infect more individuals ( and become infected more often ) than adults ., Therefore , in a procedure borrowed from 6 , we measured the age distribution of secondary cases when the index case was randomly selected and used this distribution to weight the contribution from the various age groups to the calculation to define ., The definition of applies to a population with no pre-existing immunity , an assumption that may be violated for seasonal influenza ., One can use the model to simulate seasonal influenza epidemics by substituting with the desired , the average number of people a typical infected case infects in a population with pre-existing immunity ., The simulated case generation time , or the time between infection of an individual and the transmission to susceptibles , was 3 . 4 days for a wide range of in a fully susceptible population ( Figure 2B ) ., This is consistent with other estimates for seasonal and pandemic influenza 20 , 23 ., The primary pharmaceutical intervention is vaccination ., Vaccinated individuals in the simulation have a reduced probability of becoming infected ( VES ) , of becoming ill given infection ( VEP ) , and of transmitting infection ( VEI ) 24 ., In the model , these efficacy parameters are implemented by multiplying the transmission probability per time step by ( 1−VES ) if the susceptible individual is vaccinated and by ( 1−VEI ) if the infectious individual is vaccinated ., The probability of vaccinated individuals becoming symptomatic ( ill ) after they are infected is the baseline probability ( 67% ) multiplied by ( 1−VEP ) ., Vaccines do not reach full efficacy immediately – their protective effects may gradually increase over several weeks ., The default behavior in the model is that the vaccine takes two weeks to reach maximum efficacy , with the efficacy increasing exponentially starting the day after the vaccination ., Because of the delay in reaching maximum efficacy , it may be necessary to vaccinate the population early ., In the simulation , vaccines can be administered at least four weeks before the epidemic ( i . e . , pre-vaccination ) , during the epidemic ( reactive ) , or one dose can be administered at least three weeks before the epidemic and the boost can be administered reactively ( prime-boost ) ., Antiviral agents ( neuraminidase inhibitors ) can be used for treatment of cases and for prophylaxis of susceptibles ., A single course of antiviral agents is enough for 10 days of prophylaxis or 5 days of treatment ., In the model , 5% of individuals taking antiviral agents prophylactically stop after 2 days and 5% taking them for treatment stop after 1 day 19 ., As with vaccines , individuals taking antiviral agents can have reduced susceptibility ( AVES ) , probability of becoming ill given infection ( AVEP ) , and transmitting infection ( AVEI ) ., However , unlike vaccines , the protective effects of the antiviral agents last only as long as they are being taken ( 5 to 10 days ) ., When a case is ascertained , the individual is treated with antiviral agents , and that individuals household members will also each be given a course if household targeted antiviral prophylaxis ( HHTAP ) is in effect ., Several non-pharmaceutical interventions can be simulated in the model ., School closures are simulated by eliminating school group contacts ( including preschools and daycares but not playgroups ) for those enrolled in school , but adding daytime contacts with other household members not in school or at work and doubling their daytime neighborhood and community contact probabilities to account for their non-school activities ., Schools can be closed when cases are ascertained in communities or in the schools , and they can be closed for a fixed number of days or for the duration of the simulation ., During an epidemic , individuals may be requested to stay at home if they become ill ., When simulating isolation of cases , individuals withdraw to the home one day after becoming symptomatic ( with a certain probability to represent the compliance probability ) ., This will eliminate any daytime social contacts that they have other than with household members who are not working or at school ., We simulate a liberal leave policy in a similar manner: employed individuals withdraw to the home with a pre-set compliance probability for one week one day after becoming symptomatic ., During an epidemic , those living with symptomatic individuals may be requested to stay home 25 ., In simulations of household quarantine , family members of symptomatic individuals will independently decide ( based on a compliance probability ) whether to obey quarantine for 7 days one day after the first individual becomes symptomatic ., Individuals electing to quarantine themselves withdraw to the household and interact only with household members ., If other family members become ill during quarantine , household members independently decide whether to obey quarantine for 7 days one day after each individual becomes symptomatic ., FluTE is written in C/C++ and is released under the GNU General Public License ( GPLv3 , see http://www . gnu . org/licenses/gpl . html ) ., The source code is available at http://www . csquid . org/software , https://www . epimodels . org/midas/flute . do , and the Models of Infectious Disease Agent Study ( MIDAS ) repository 26 ., The software includes two source code files that are also freely distributable but may come with different licenses because they were written by others: one for the pseudorandom number generator ( SIMD oriented Fast Mersenne Twister ( SFMT ) pseudorandom number generator 27 ) and one to generate binomially distributed random numbers ( from Numerical Recipes in C 28 ) ., Version 1 . 11 of FluTE was used to produce the results in this manuscript ., A configuration file is used to specify the population to use for the simulation , the parameters for starting the epidemic , the transmissibility of the infectious agent , and the desired intervention strategies ., The configuration file is text-based and can be typed in by a user or generated with a script ., The simulation outputs results to text files , which can be easily parsed for plotting or statistical analysis ., A parallelized version of the code supports simulations of large populations ( up to the entire continental United States ) ., This version of the program assigns the populations of different counties to different processors , and OpenMPI is used to update the status of individuals who travel between communities that are located on different processors and to update the global status of the epidemic and the interventions ( e . g . , the total number of vaccines used ) ., The simulation uses approximately 80 megabytes of memory per million simulated individuals ., The simulation was written with several competing goals: to explicitly represent each individual in the population , to conserve memory , to run quickly , and to be ( relatively ) easy to read and modify ., Each simulated individual is represented by a C structure that includes unique identifiers for the person and for each of the social mixing groups to which that person belongs , the age of the individual , the persons infection and vaccination status and dates , and other attributes ., For each infected individual , the simulation identifies all susceptible individuals in that persons community who share a common mixing group , the infectiousness of the infected individual , the susceptibility of the susceptible , and the probability that transmission takes place for every time step ., Although comparing each individual with every other within a community results in the number of comparisons increasing with the square of the number of individuals , community sizes are always smaller than 3 , 000 residents ., Therefore , the number of comparisons made between individuals scales approximately linearly with the number of individuals in the simulation ., More sophisticated algorithms could improve the simulations performance , but may do so at the expense of the codes flexibility and readability ., The running time depends on the number of individuals infected during the course of a simulation ., Simulating an epidemic in a population of 10 million people can take up to two hours ( on a single processor on an Intel Core2 Duo T9400 ) , but it may take only seconds if the virus is not highly transmissible ( low ) or if there are effective interventions ( e . g . , high vaccination rates ) ., On a cluster of 32 processors , simulating an epidemic covering the continental United States ( population of 280 million ) takes about 6 hours ( 192 hours of total CPU time ) ., We illustrate the use of the model by simulating epidemics in metropolitan Seattle , a major metropolitan area with a population of approximately 560 , 000 according to the US 2000 Census ., We ran simulations with different values of , starting with ten infected individuals chosen at random , and found that the epidemic could peak as early as 45 days after the start if is high ( ) ( Figure 3A ) ., Pre-vaccination ( with vaccine efficacies of VES\u200a=\u200a40% , VEP\u200a=\u200a67% , VEI\u200a=\u200a40% , which correspond to a well-matched seasonal influenza vaccine 29 ) is likely to both lower and delay the epidemic peak ( Figure 3B ) ., Use of antivirals alone ( AVES\u200a=\u200a30% , AVEP\u200a=\u200a60% , and AVEI\u200a=\u200a62% 11 ) did not greatly reduce the epidemic peak , but they could reduce illness and mortality in an epidemic ., Non-pharmaceutical interventions could be quite effective , but the epidemic may spike immediately upon ending the intervention ( compare permanent school closure with school closure for 60 days in Figure 3B ) ., The illness attack rates in the simulation are lower than those in a SIR model with random mixing ( where 30 , where AR is the infection attack rate , and the illness attack rate is 0 . 67AR ) ( Figure 3C ) ., As observed in earlier studies , models with community structure have lower attack rates than those with random mixing 31–33 ., Simulated epidemics struck school-age children earlier than adults , which had been observed in earlier studies 6 , 34 ., Therefore , we predict that early in an epidemic , the proportion of cases who are school-age children will be higher than later in the epidemic ( Figure 4 ) ., This phenomenon might affect the accuracy of estimates in unfolding epidemics ., For example , most confirmed cases in the recent novel influenza A ( H1N1 ) outbreaks in the United States have been school-age children 35 and several early estimates of have been above 2 36 , 37 ., In our model , we observed that infected children generate more secondary cases than infected adults ( Figure 2A ) ., For example , infected school-age children would transmit to an average of other individuals in a simulated epidemic with ., Therefore , estimates of could be high early in an epidemic when a disproportionate number of infections are in children ., One can simulate the population of the entire continental US using the parallel version of FluTE ( mpiflute ) ., The continental US had 280 million people in 64735 census tracts in 2000 , based on the US 2000 Census ., In our simulations , we found that the final illness attack rates for the US to be nearly identical to those of metropolitan Seattle , but the epidemic peak for a given is later for the United States ( e . g . , 94 vs 65 days for ) ( Figure 5 ) ., Therefore , simulations of a sufficiently large metropolitan area may be adequate for determining the effect of a strategy on the national level on final illness attack rates , but the nation-wide peak of the epidemic may be later than in the major metropolitan areas because of the time it takes the epidemic to reach outlying areas ., We have described a new publicly available influenza epidemic simulator , FluTE ., It explicitly represents every individual in the simulation , so simulated epidemics can be studied in detail , even tracing individual transmission events ., We illustrated the use of FluTE with examples in which we explored the effect of various intervention strategies on influenza epidemics in the United States and showed how transmissibility can be over-estimated early in an epidemic ., The simulation was written so that one can easily set the transmissibility , vaccination policies ( e . g . , fraction of the population to vaccinate ) , and other reactive strategies ( e . g . , school closures ) ., These settings can be used to investigate questions such as:, 1 ) What fraction of the population will become infected or ill ?, 2 ) How much vaccine coverage is required to mitigate an epidemic with a given ?, 3 ) What segment of the population should be vaccinated to reduce overall illness attack rates the most ?, 4 ) How long can one wait before reacting to an epidemic ?, and, 5 ) What range of can be managed by a particular pandemic strategy ?, We have used FluTE to investigate some of these questions by simulating vaccinating children against seasonal and pandemic influenza 38 and pandemic mitigation 20 ., The model was calibrated to simulate epidemics of a virus similar to 1957/1958 Asian A ( H2N2 ) and 2009 pandemic A ( H1N1 ) ., We attempted to model realistic pharmaceutical and non-pharmaceutical interventions , but their effects on an epidemic have not been well quantified ., The models results are plausible and likely to be qualitatively correct , but there is insufficient data to calibrate it to produce quantitatively accurate results for the various possible disease parameters and mitigation strategies ., Although the model generates realistic population-level results , the spatial dynamics of the epidemics it produces should be used for illustrative purposes only ., When using the model to evaluate mitigation strategies , it is important to consider ones goals ., For example , using antiviral agents to treat cases does not greatly reduce the final illness attack rate in the simulation , but it could greatly reduce mortality ., The model does not directly evaluate the cost of interventions , but the numbers of cases in a simulated epidemic can be linked to cost and healthcare utilization data 39 ., Differential equation models are the most popular approach to disease modeling ., The simplest of these ( such as the SIR model 40 ) can be used to study epidemics analytically , and more complex versions have been used to model the dynamics of epidemics on a global scale 41 , 42 ., However , if one wants to include a complicated natural history of disease or detailed intervention strategies , individual-based models , such as FluTE , may be more suitable ., The current software supports a limited set of configuration options and is intended for batch runs using a scripting language ., Using the model for scenarios not supported by the existing code , such as testing a novel intervention strategy or altering the contact parameters for a different attack rate pattern , would require modification of the source code , which we have released so that others can make such changes if needed ., We decided to adopt the GNU General Public License ( GPL ) , so that the source code of derivative works must be released ., We believe this will facilitate the sharing of improvements ., The availability of source code allows others to adapt the model to simulate outbreaks of other airborne infectious diseases such as smallpox 3 , 43 , 44 or to simulate other regions of the world with different social structures 3 ., In the future , we would like to make our model more accessible to non-programmers ., This may involve developing a user interface or adding new parameters to the configuration file ., We would also like to include intervention strategies that best reflect government pandemic mitigation plans ., Achieving these goals would depend upon close collaboration with public health officials to better understand their needs and to carefully simulate existing pandemic mitigation plans and capacities ., Although we have calibrated our model to the best available data , more detailed and reliable information on the natural history of influenza , influenza transmission , human behavior in response to infection , and vaccine efficacy is needed ., Sensitivity analyses of similar epidemic models have shown that results are robust to uncertainty in many parameters 3 , 5 , 6 , 11 ., However , more accurate model inputs would improve the quantitative predictions ., Well-designed studies are needed to acquire these data .

Introduction, Model, Results, Discussion

Mathematical and computer models of epidemics have contributed to our understanding of the spread of infectious disease and the measures needed to contain or mitigate them ., To help prepare for future influenza seasonal epidemics or pandemics , we developed a new stochastic model of the spread of influenza across a large population ., Individuals in this model have realistic social contact networks , and transmission and infections are based on the current state of knowledge of the natural history of influenza ., The model has been calibrated so that outcomes are consistent with the 1957/1958 Asian A ( H2N2 ) and 2009 pandemic A ( H1N1 ) influenza viruses ., We present examples of how this model can be used to study the dynamics of influenza epidemics in the United States and simulate how to mitigate or delay them using pharmaceutical interventions and social distancing measures ., Computer simulation models play an essential role in informing public policy and evaluating pandemic preparedness plans ., We have made the source code of this model publicly available to encourage its use and further development .

Computer simulations can provide valuable information to communities preparing for epidemics ., These simulations can be used to investigate the effectiveness of various intervention strategies in reducing or delaying the peak of an epidemic ., We have made a detailed influenza epidemic simulator for the United States publicly available so that others may use the software to inform public policy or adapt it to suit their needs .

public health and epidemiology/epidemiology, public health and epidemiology/infectious diseases, infectious diseases/epidemiology and control of infectious diseases, computational biology

journal.pntd.0001946

Leishmaniasis Direct Agglutination Test: Using Pictorials as Training Materials to Reduce Inter-Reader Variability and Improve Accuracy

Up until the 1990s accurate visceral leishmaniasis ( VL ) diagnosis necessitated parasitological confirmation by microscopy or culture of the blood , bone-marrow , lymph nodes or spleen 1 ., Microscopic detection of parasites in clinical material from the spleen is still considered the reference standard; however , splenic aspirates are associated with risk of serious bleeding and should only be carried out in settings with access to blood transfusion and surgical services ., The invasiveness and potentially fatal complications associated with splenic aspiration has spurred the development of non-invasive serological tests such as direct agglutination test ( DAT ) 2 over 25 years ago and in the past decade , lateral flow immuno-chromatographic tests ( ICT ) , commonly referred to as rapid diagnostic tests ( RDTs ) ., RDTs have now been adopted widely , in the Indian subcontinent 3 , but in other endemic regions , DAT is part of the diagnostic algorithm or is used for epidemiological surveys due to variable sensitivity of RDTs 2 , 4 ., The DAT , in its present form , is a freeze dried suspension of trypsin-treated fixed and stained culture of L . donovani promastigotes 5; liquid formulations of DAT are also manufactured locally ., During infection with VL , circulating antibodies are produced against the surface antigens of the invading parasites ., The DAT detects antibodies to L . donovani s . l . in the blood or serum of those infected by means of direct agglutination ., In the absence of antibodies to Leishmania the DAT antigen accumulates at the bottom of the plate to form a dark blue spot ., If antibodies to Leishmania are present then the antigen forms a pale blue film over the well and this constitutes a positive result ., DAT requires moderate technical expertise , and laboratory equipment and reagents , including calibrated pipettes , micro-titre plates , multiple reagents and a toxic solution ( chemical 2-beta Mercapto-ethanol ( 2-ME ) ) 2 ., Furthermore , despite very good accuracy , inter-observer discrepancy in routine DAT serology readings is common 6–8 ., Prompted by shared experiences of six endemic countries using DAT to characterize performance panel samples , we report an assessment of DAT inter-reader variability ., It was noted that the inter-laboratory agreement of DAT titres on a panel of 15 sera was low ., Here , our objective was to standardize the reading of DAT by developing and implementing pictorial training aids ., Nine laboratories from three global endemic regions were involved in a WHO/TDR-sponsored evaluation of VL RDTs; namely Asia ( n\u200a=\u200a4 ) , South America ( n\u200a=\u200a2 ) and Eastern Africa ( n\u200a=\u200a3 ) ( Table 1 ) ., The Institute of Tropical Medicine , Antwerp , Belgium ( ITM ) assembled a proficiency panel including sera from 10 VL confirmed patients including a range of DAT titres , and 5 VL negative patients , one healthy endemic control and others who harbored potentially cross-reacting , infections , including Chagas disease , tuberculosis , malaria and leprosy ., Prior to shipping , each sample within the panel was assigned a random numerical code that varied from centre to centre ., All samples were left over from samples that had been taken as part of research projects conducted between 1978 and 2000 at the Institute for Tropical Medicine ( ITM ) Antwerp , Belgium ., The samples were anonymised and kept stored for future use for scientific purposes ., In the studies conducted since 2000 explicit consent was asked for storage and future use of left overs of the samples that were taken ., In the older studies no explicit mention was made of future use of the stored left overs though a general informed consent was asked ., However , it was not possible to trace back the study participants in the studies preceding 2000 and to ask them for informed consent for storage and use of left over samples The proficiency panel was tested blindly using the DAT assay ( KIT-Biomedical Research , Lot 0904 ) in each of the nine evaluation laboratories ( Table 1 ) and both reference laboratories ( KIT and ITM ) ., Results were returned electronically to ITM using a standard recording form ., All microtitre plates used in the procedure were provided by reference laboratories ( Greiner 651101 100 ) ., The DAT was performed as described previously 2 ., Due to significant discordance in end-titres between all laboratories , photographs of DAT plates with 10 VL positive serum samples and 5 VL negative serum samples were prepared by the reference laboratories ( following joint agreement on end titres ) and were used as pictorial training aids ., Refresher DAT training was given by staff of KIT and Banaras Hindu University ( BHU ) to all participating laboratories ( Table 1 ) ., Trainers assessed equipment and compliance with the DAT SOP , including preparation of reagents ., Subsequently , the proficiency panel was repeated in the presence of the trainer ., End titres were read independently by two separate technicians and the trainer ., When readers did not agree on the end titre they came to a common conclusion after joint discussion ., Combined results of the readers were sent to ITM and decoded by a study team member not involved in the refresher training; results of the trainer were not taken into consideration unless the results of the readers were significantly different from those of the reference laboratories and the test was repeated ., Disagreement was defined as greater than one titre above or below those of the reference laboratories 6 ., Variance in results before and after refresher training was compared with a two-sample Wilcoxon rank-sum ( Mann-Whitney ) test ., Despite having received the same panels , batch of DAT , microtitre plates and protocol , overall DAT results concordance ( agreement within one titre ) with the reference laboratories was only 50% ., Agreement on negative controls was very good ( 94% ) ., Using a cut off of 1∶1600 serum dilution , the pre-training sensitivity and specificity were 79% and 94% , respectively ., Refresher training was initiated due to the large differences in DAT reading between participating laboratories ., Here , photographs of DAT plates were used as training aids , where end-titres had been agreed upon by the reference laboratories ., During refresher training the trainers did not identify any faulty or inappropriate equipment , nor did they witness any non-compliance with the DAT SOP ., After refresher training the concordance ( agreement with one DAT titre ) increased to 84% with the reference laboratories ., The agreement on negative controls increased to 98% ., Average variance in results before refresher training was 3 . 3 titres; this improved to an average variance of 1 . 0 titre reading ( the accepted limit ) after refresher training ., A non-parametric test was used to test for significant differences before and after training using a two-sample Wilcoxon rank-sum ( Mann-Whitney ) which showed significant difference ( z\u200a=\u200a−3 , 624 and p\u200a=\u200a0 . 0003 ) ., Post-training the sensitivity increased to 97% and the specificity to 100% ( cut off values 1∶1600 ) ., Overall , the refresher training increased the operator performance of the DAT in this small proficiency panel ( Table 2 ) ., After refresher training a cut-off point of 1∶1 , 600 ( serum dilution ) gave 97% sensitivity ( CI: 91 . 6–99 . 0% ) and 100% specificity ., Further , pictorial guides ( Figures 1 , 2 , and 3 ) of DAT training plates reflecting consensus end titres by several experts in the VL-LN , with many years of experience in using DAT as a diagnostic tool , are now available ., Further recommendations to be taken into account for completion of the DAT assay are highlighted in Table 3 and data per laboratory pre-training and post-training with pictorial aids can be seen in Supporting Information S1 ., The DAT assay has been used as a diagnostic tool for more than 25 years , it is robust , reliable has a high clinical accuracy and can be performed in laboratories with minimal equipment ., However , the subjective manner in which the result ( end-titre ) of the test is read means that inter-reader variation in titre reading can be an issue ., Preparations for a multicenter evaluation of RDTs unexpectedly uncovered a significant discordance in DAT results among reference and evaluation centres; this presented an opportunity to address discordance and create an international , consensus-based protocol and training materials to strengthen standardized reading of the DAT for VL diagnosis without compromising diagnostic accuracy ., The reasons for all of the discrepancies between the different laboratories is not fully understood however , it was noted that technicians were generally competent in the DAT procedure , particularly those who used it as part of their routine diagnostic algorithm ., It was not possible to test the saline that the laboratories previously used in testing , but it is possible that the origin and quality of the saline solutions used as a diluent for the DAT antigen did affect performance , generating false positive precipitation in negative sera wells ., The most consistent problem identified in the laboratories can be attributed to the subjective manner in which the end titre of the DAT test is typically read ., Some readers record the end-titre when 50% of agglutination of the well has occurred ( as occurs with other agglutination tests ) , whilst other readers record the end-titre where the whole well has agglutinated and there is no difference between a negative control well ( antigen plus saline ) and the sample well ., Even though 1 titre difference in reading is considered acceptable 6 , the discrepancy and variance in results reported here was far greater ., Training plates developed by the reference laboratories proved to be extremely helpful in illustrating the end titre ., Positive sample wells were defined by any reaction in the test well in comparison to the negative control well; this ensured that the high sensitivity of the DAT was not compromised ., Figure 2 shows the end-titre as agreed by the VL-LN; a follow up plate can be used to test users before revealing the results as seen in figure 3 ., High quality photos in figures 2 and 3 are also available by request ( contact corresponding author ) for use as reference training material for future DAT users ., Since slight variations in readings between different DAT antigen batches may occur it is advised that the same batch of DAT should be used within one project or epidemic to decrease variability in results ., If this is not possible then it is recommended to keep reference sera in order to assess this lot-to-lot variation , this should not be more than one titre difference ., In addition , it is important that all users of the DAT specify the type of dilution used , i . e . serum dilution ( starting 1∶100 ) or antigen plus serum dilution ( starting 1∶200 ) ., It is likely that cut-off values are different between endemic areas and even during epidemic cycles ., Local guidance as to appropriate cut-off values is essential ., The problems uncovered during a multicenter DAT proficiency testing scheme are potentially relevant to other DAT users ., To reduce inter-reader variability and increase accuracy , photos of training plates were made , and end-titres were agreed upon firstly by the reference laboratories and subsequently by experienced users of DAT within the VL-LN ., These photos can be used to promote a standardized approach to interpreting DAT without compromising sensitivity ., Protocols and photos can be requested for training and quality control purposes by two of the major manufacturers of the assay , KIT and ITM ., High sensitivity and specificity can be achieved with this reliable and robust diagnostic tool , and we hope that provision of good training materials can increase the usefulness of DAT .

Introduction, Materials and Methods, Results, Discussion

The Direct Agglutination Test ( DAT ) has a high diagnostic accuracy and remains , in some geographical areas , part of the diagnostic algorithm for Visceral Leishmaniasis ( VL ) ., However , subjective interpretation of results introduces potential for inter-reader variation ., We report an assessment of inter-laboratory agreement and propose a pictorial-based approach to standardize reading of the DAT ., In preparation for a comparative evaluation of immunochromatographic diagnostics for VL , a proficiency panel of 15 well-characterized sera , DAT-antigen from a single batch and common protocol was sent to nine laboratories in Latin-America , East-Africa and Asia ., Agreement ( i . e . , equal titre or within 1 titer ) with the reading by the reference laboratory was computed ., Due to significant inter-laboratory disagreement on-site refresher training was provided to all technicians performing DAT ., Photos of training plates were made , and end-titres agreed upon by experienced users of DAT within the Visceral-Leishmaniasis Laboratory-Network ( VL-LN ) ., Pre-training , concordance in DAT results with reference laboratories was only 50% , although agreement on negative sera was high ( 94% ) ., After refresher training concordance increased to 84%; agreement on negative controls increased to 98% ., Variance in readings significantly decreased after training from 3 . 3 titres to an average of 1 . 0 titre ( two-sample Wilcoxon rank-sum ( Mann-Whitney ) test ( z\u200a=\u200a−3 , 624 and p\u200a=\u200a0 . 0003 ) ) ., The most probable explanation for disagreement was subjective endpoint reading ., Using pictorials as training materials may be a useful tool to reduce disparity in results and promote more standardized reading of DAT , without compromising diagnostic sensitivity .

Until the 1990s accurate Visceral Leishmaniasis ( VL ) diagnosis necessitated parasitological confirmation by microscopy or culture of the blood , bone-marrow , lymph nodes or spleen ., These techniques are invasive and splenic aspirates are associated with a risk of serious bleeding ., This has led to the development of non-invasive serological tests such as the direct agglutination test ( DAT ) ., During infection with VL , circulating antibodies are produced against the surface antigens of the invading parasites ., The DAT detects antibodies to L . donovani s . l . in the blood or serum of those infected by means of direct agglutination ., In the absence of antibodies to Leishmania the DAT antigen accumulates at the bottom of the plate to form a dark blue spot ., If antibodies to Leishmania are present then the antigen forms a pale blue film over the well constituting a positive result ., Here , we report on shared experiences of six endemic countries using DAT to characterize performance panel samples ., There was considerable inter-reader variability and in order to standardize the reading of DAT we developed and implemented pictorial training aids ., After refresher training , agreement between readers increased; the pictorial aids and recommendations for using DAT are available in this article .

medicine, infectious diseases, test evaluation, diagnostic medicine, leishmaniasis, protozoan infections, parasitic diseases

journal.ppat.1003104

The Roles of Competition and Mutation in Shaping Antigenic and Genetic Diversity in Influenza

Influenza viruses are classified into types A–C , among which influenza A is the most pathogenic ., These viruses cause between a quarter to half a million deaths worldwide 1 and tens of thousands of deaths in the US during annual epidemics 2 ., The economic burden of seasonal influenza in the US is estimated at more than ten billion dollars in healthcare costs alone 3 ., The major targets of humoral immunity against influenza A are its envelope glycoproteins , hemagglutinin ( HA ) and neuraminidase ( NA ) ; these form the basis of its crude classification into subtypes H1N1 , H2N2 and H3N2 etc ., Since its emergence in 1968 , influenza A ( H3N2 ) has continually circulated in the human population ., The phylogeny of its HA protein ( Figure 1 ) shows a distinctive ‘cactus-like’ shape with a narrow , usually single-trunked , tree 4 , ., The ‘narrowness’ of the tree is derived from the fact that contemporaneous H3 proteins share a single common ancestor 2–8 years in the past 6 , 7 ., This short time is unique to H3N2 given its global spread and its high prevalence and incidence 7 ., The classical view of influenza evolution is one of antigenic drift 8 , 9 , 10 in which antigenic change continually and gradually accumulates in the virus through the influence of selection by way of changes to the HA and NA proteins ., By itself , the ‘cactus-like’ structure of the A/H3N2 phylogenetic tree suggests the presence of adaptive evolution 7 and several studies have provided evidence for positive selection 5 , 11 , 12 , 13 , 14 ., However , it is difficult to explain the limited standing diversity of influenza 15 , 16 , and the empirical evidence for discontinuous antigenic change 17 , 18 , under a general antigenic drift framework ., Multiple epidemiological hypotheses have been advanced to reconcile the these observations with a process of continual antigenic divergence including short-lived strain-transcending immunity 15 , 16 , 19 , epochal or punctuated evolution 13 , 17 , trait-space reduction 20 and canalized evolution 21 ., A competing hypothesis advanced by Recker et al . 22 eschews the paradigm of antigenic drift , instead considering that , owing to functional constraints on the defining epitopes , the virus population is limited phenotypically to a restricted set of antigenic types ., Antigenic types replace each other with waves of dominance resulting from frequency-dependent immune mediated selection as “niches” in antigenic space are dynamically generated and are exploited by the existing virus population ., In its original implementation , this model assumes that all antigenic types remain present in the population in low frequencies , as an approximation to the idea that they can be generated by mutation from preexisting strains at a sufficient rate as not to limit the emergence of a type favored by selection ., Thus , the model describes in practice a case where influenza outbreaks are caused by host immune selection in a manner that is not limited by the rate of antigenic mutations ., Although patterns of turnover are consistent with those observed for H3N2 , it is not clear whether the characteristic phylogenetic trees can be generated by this model ., Here , we have attempted to resolve this question using a large-scale individual-based simulation of epidemiological and evolutionary dynamics that allows the complete phylogenetic tracking of a virus population characterized by defined repertoires of polymorphic epitopes ., Our model is based on the multi-locus structure employed by 22 with host immunity operating at an epitope-specific level ., When contacted by a virus , a hosts risk of infection is determined by the number of alleles/epitopes recognized by its immune system ., We also introduce the possibility of a long-lived strain-transcending component to the model ., Thus , competition between strains is determined both by the number of shared epitopes and a variable level of generalized immunity ., Our model differs in this regard from that of Recker et al . 22 which does not permit full cross-protection except in the case of having experienced the exact same combination of epitopes , a feature that implicitly accounts for the effect of highly variable epitopes unique to each strain ., This model structure allows us to make inferences about the roles of mutation and competition in a more general context ., Models of antigenic dynamics tend to polarize between those in which the availability of antigenic types dictate the dynamics 13 , 17 , 23 , 24 , and those where host immune-mediated selection is the only driver 25 , 26 ., We refer to the latter regime , where antigenic change is constrained by host population immunity , as selection limited , whereas the former , in which the availability of antigenic mutations pose the rate limiting step , is described as mutation limited ., The approach we take in this paper offers a tool for locating influenza on this continuum and would easily generalize to other antigenically diverse infectious diseases ., To explore the epidemiological dynamics of our model in isolation , we implemented a parameterization lacking mutation , in which extinction was preempted by maintaining at least one carrier for each antigenic-phenotype ., Different colors ( Figure 3 ) represent the prevalence of different antigenic-phenotypes ., Here , the antigenic repertoire is derived from combinations of variants at 5 distinct epitopes ( see Methods for full description of epidemiological parameters ) ., A mutation-free model can result in different alternative dynamics based on model parameters ranging from stable coexistence of completely discordant antigenic-phenotypes to the successive replacement of strains through chaotic or cyclic behavior 25 ., Not surprisingly , our model implementation with no explicit evolution also generates these waves of replacement ( Figure 3 ) , suggestive of H3N2 influenza as proposed by Recker et al . 22 ., We examined the effects of mutation at different rates on the resulting phylodynamic patterns of the virus by seeding the population with a single strain and tracking antigenic and evolutionary changes ., We measure diversity π as the average time separating two randomly selected contemporaneous viruses since their divergence from a common ancestor ., Because branch lengths in our genealogies are measured in years , the resulting diversity is also measured in years ., In the absence of antigenic mutation , only a single strain persists , experiencing transient oscillatory dynamics between near extinction , and endemic equilibrium conditions ( Figure 4A ) ., As all viral traits are equal , there are no selective forces and the observed phylogeny and coalescence rates can be directly related to prevalence and incidence 29 ., This yields random coalescence within contemporaneous viral lineages and a large associated pairwise genetic diversity ( π\u200a=\u200a30±12 years; mean±std across 5 simulations ) ., For low mutation rates ( Figure 4B ) the introduction of new mutations is the critical determinant of strain dynamics ., Each new variant outcompetes the one that came before , resulting in a spindly phylogenetic tree and therefore low diversity ( π\u200a=\u200a5 . 7±0 . 8 years ) ., Temporally adjacent strains are antigenically similar , rather than discordant , forcing strong competitive exclusion and single strain dominance ( ε\u200a=\u200a0 . 93±0 . 02 ) , where ε is the proportion of infections caused by the most common strain ., An increase in mutation rate ( Figure 4C ) leads to deeper branches with a corresponding increase in phylogenetic diversity ( π\u200a=\u200a220±100 years ) and more pronounced antigenic divergence ., Here , the population dynamics are ruled by the endemic or cyclic behavior of discordant antigenic sets ., The emergence of new intermediate antigenic types is suppressed by competition from the two prevalent strains 25 ., At a relatively high mutation rate ( Figure 4D ) , we approach population dynamics similar in appearance to those of the mutation-free model ( Figure 3 ) ., Diversity is high ( π\u200a=\u200a120±30 years ) , with deep yet occasionally coalescing branches ., In general , a threshold exists at which mutation overwhelms selection resulting in a population drifting away from the fittest genotype 30 ., For sufficiently high rates of antigenic mutation , all antigenic types reach near equal frequency in the population ( Figure 4E ) ., On a population scale , the high mutation rate weakens frequency-dependent selection and results in the breakage of antigenic strain structure; antigenic types do not cluster across the genealogy ., The loss of selection forces breaks down phylogenetic structure and leads to a reduction in the depth of the branches ( π\u200a=\u200a120±70 years ) compared to the one observed for discordant antigenic types ., By comparing fixation versus extinction of antigenic mutations , using a quantity related to the McDonald-Kreitman ( MK ) index 31 , 32 , 33 we estimated the strength and direction of selection on antigenic mutations in our model ( see Methods ) ., Here we calculate an MK related ( MKR ) index as the ratio of the per-year rate of antigenic mutation on the trunk to the per-year rate of antigenic mutation on the side branches ., If antigenic mutations are advantageous for long-term virus persistence , an MKR ratio above 1 is expected ., In this case , individuals exhibiting these antigenic mutations will be more likely to fix in the population and contribute to substitutions on the trunk of the phylogeny ., Similarly , if antigenic mutations are deleterious to the long-term success of the virus , an MKR index of less than 1 is expected ., This is because mutant individuals will tend to be lost from the population and side branches will show an excess rate of substitution ., We find that , when rare , antigenic mutations show highly increased rates of fixation ( MKR\u200a=\u200a19±11 ) , and therefore evidence of strong positive selection ( Figure 4B ) ., Hence , we find that strong positive selection results in both a spindly tree and an overabundance of antigenic mutations of the trunk of the phylogeny ., An increase in the mutation rate leads to the emergence of antigenically discordant types , and the suppression of other antigenic mutants ( Figure 4C ) ; here , we find strong negative selection mediated by host immunity with an MKR index of 0 . 47±0 . 22 ., At a still higher mutation rate ( Figure 4D ) , we observe a balance of positive and negative selection resulting in MKR\u200a=\u200a1 . 1±0 . 3 ., With saturating mutation rates ( Figure 4E ) we further lose the signature of selection ( MKR\u200a=\u200a1 . 0±0 . 1 ) on phylodynamic patterns ., Our model contains a cross-immunity parameter σ which allows us to explore a range of immune selection regimes: when σ\u200a=\u200a1 , we have full cross-protection ( as might arise if each epitope elicited a very strong immune response ) and when σ\u200a=\u200a0 , cross-protection between strains is only high if they share their entire variable repertoire ., In general , stronger cross-immunity results in lower prevalence as hosts fail to be re-infected ( Figure S1 ) ., We find that , for most of the parameter space , genetic ( genealogical ) diversity π , increases with weaker cross-immunity and with more rapid mutation ( Figure 5A ) ., The ( mostly ) monotonic relationship between competition and diversity is broken at the threshold of limiting similarity 34 where , regardless of epitope differences , two strains suffer full cross-protection ., This scenario , shown as a band on the right-hand side of Figure 5A where σ\u200a=\u200a1 , results in the disappearance of selective effects and greater levels of genetic diversity ., Here , diversity rebounds to its neutral expectation due to random coalescence ., Exceptions to the monotonic pattern of diversity with competition can also be found for intermediate mutation rates ., The relationship between mutation , cross-immunity , and the MKR index is less straight-forward ( Figure 5B ) ., Here , the highest levels of positive selection are present when cross-immunity is strong ( σ\u200a=\u200a0 . 8–0 . 9 ) , and mutation is weak ( ξ\u200a=\u200a10−5 ) ., When mutation rate is limiting ( ξ<5×10−5 ) then antigenic mutations are favored by natural selection ( MKR>1 ) ., However , when mutation rates are higher ( 5×10−5<ξ<10−3 ) , negative selection by-and-large predominates ., The strongest negative selection occurs in a region of moderate cross-immunity ( σ\u200a=\u200a0 . 6 ) corresponding to previously observed discordant dynamics ( Figure 4C ) ., There is also a clear relationship between diversity and selection as measured by the MKR index ( Figure 5C ) ., We observe a strong negative correlation between MKR and levels of diversity ( ρ\u200a=\u200a−0 . 86; Pearsons correlation ) ., If we separate results into a regime of positive selection ( MKR>1 ) and a regime of negative selection ( MKR<1 ) , we observe similar results within each regime ., Stronger positive selection coincides with a decrease in genetic diversity ( ρ\u200a=\u200a−0 . 85 when MKR>1 ) , and stronger negative selection tends further increase diversity through the persistence of discordant strains and associated deep branches ( ρ\u200a=\u200a−0 . 28 when MKR<1 ) ., As expected from population genetic theory 29 , increases in viral prevalence also coincide with increases in viral diversity , however , the correlation is weaker under positive selection ( ρ\u200a=\u200a0 . 74 when MKR>1 ) than under negative selection ( ρ\u200a=\u200a0 . 87 when MKR<1 ) and cannot be trivially dissociated from the effects of selection ., Two additional strain diversity measurements based on the ecological dynamics are shown in Figure S2 , the Shannon diversity index and the level of single strain dominance ( Methods ) ., Similar to genetic diversity , positive selection is correlated with an increase in single strain dominance ( ρ\u200a=\u200a0 . 88 when MKR>1 ) and a decrease in Shannon diversity ( ρ\u200a=\u200a−0 . 88 when MKR>1 ) ., Negative selection decreases Shannon diversity ( ρ\u200a=\u200a0 . 39 when MKR<1 ) and increases single strain dominance ( ρ\u200a=\u200a−0 . 23 when MKR<1 ) ., While negative selection lowers the number of circulating strains , it increases genetic diversity π through the existence of deep non coalescing branches ., The patterns described so far suggest that the dynamics of H3N2 influenza within this framework correspond to a regime in which host immune mediated selection is strong and the antigenic mutation rate is low ., We now extend the model in order to examine other characteristics relevant to H3N2 in a more detailed epidemiological setting that includes seasonality and a basic global population structure ., In this analysis we include three demes representing the northern hemisphere , the southern hemisphere and the tropics ., Northern and southern hemisphere demes experience an opposing seasonal modulation ( with a 14% amplitude and six months phase difference ) while tropical regions experience two weaker seasons annually 35 ( see Figure S3 and Methods ) ., In addition the southern hemisphere population is reduced in comparison to northern hemisphere and tropical populations ( Methods ) ., In this model we use an antigenic repertoire with 4 epitopes differing in the number of alternative variants per epitope ., A typical tree for this configuration together with the corresponding diversity skyline is depicted in Figure 6A ., We observe 13±6 antigenic clusters that come to dominate the virus population over the course of the 40 year simulation ( Figure S4-A ) with an average duration of 4±2 years ., Clusters are defined based on cumulative changes in two or more epitopes based on 36 ( Methods ) ., The turnover of virus strains results in a characteristic spindly phylogenetic tree and low standing genetic diversity ( π\u200a=\u200a5 . 7±0 . 1 years ) ., Over the course of the 40-year timespan , genetic diversity experiences a boom and bust pattern ( Figure 6A ) with a 10%–90% range of 3–9 . 5 years measured by combining diversity skylines of five repeated simulations ., The repeat of exact antigenic types is uncommon in the model ( Figure S5 ) while epitopes with more restricted variability ( 2–3 variants ) frequently reemerge ( Figure S6 ) ., Average yearly incidence in the northern and southern hemisphere demes is ( 5 . 7%±0 . 1 , 5 . 8%±0 . 1 ) respectively ( Figure 6B ) , while incidence in the tropics is slightly lower 5 . 5%±0 . 1 ., Annual epidemics are generated almost regularly yet display a high level of variability in peak size in the northern hemisphere ( CoV\u200a=\u200a1 . 1±0 . 1 , coefficient of variation ) and lower variability in the tropics ( CoV\u200a=\u200a0 . 7±0 . 1 ) ., The interquartile range in peak weekly cases ranges ( IQR≈200–800 cases per 100000 ) in the northern hemisphere and ( IQR≈400–1100 ) in the tropics ., In this model the tropics or lower and mixed seasonality populations exhibit a greater role ( 68%±9 ) in establishing the trunk of the influenza tree ( Figure 6B ) ., The southern hemisphere experiences a smaller ( 12%±3 ) part in establishing the trunk of the tree in comparison to the northern hemisphere ( 20%±8 ) ., In addition we find that antigenic variants are more likely to reach significant prevalence in the tropics earlier making the tropics “antigenically ahead” ( Figure S4-B ) ., Antigenic variants reach 5% of their total deme prevalence 2±1 . 5 months earlier in the tropics compared to the northern hemisphere and 3±2 months earlier in the tropics compared to the southern hemisphere ( p<0 . 001 for the combined results ) ., Similarly antigenic variants decline ( reach 95% of their total prevalence ) earlier ( 1 . 7±0 . 3 month , p<0 . 0005 ) in the tropics compared to the northern hemisphere , yet not significantly earlier or later than the southern hemisphere ., We find that strong competition and high R0 values generates more regular annual epidemic peaks while maintaining low genetic diversity ., In addition we find that within antigenic cluster evolution also contributes to maintain low genetic diversity ., An increase ( 0 . 005 compared to 0 . 001 of contacts ) in the strength of the metapopulation coupling slightly improves the epidemiology by decreasing the likelihood of long periods without annual epidemics ., To establish whether low genetic diversity can be maintained when the number of epitopes or variants per epitope is increased we repeated the same parameterization with double the number of epitopes and with twice the number of variants per epitope and by keeping either the per-epitope or overall mutation rate ., The results are summarized in Table S1 ., When doubling the number of epitopes , the model can attain similar results with respect to genetic diversity π and overall incidence , when the total mutation rate across epitopes is maintained and the cross-immunity decrease per–epitope change is halved ., In contrast , it is not clear whether a model that includes an increase in the number of variants per epitope can maintain low genetic diversity levels and maintain similar or higher incidence levels ., Herein , we implemented an individual-based model that allowed us to track both the ecological and evolutionary dynamics of a pathogen population , in which cross-immunity is orchestrated by a finite set of antigenic loci of limited variability 22 ., We used this model to compare phylodynamic patterns under a regime governed primarily by limitation on the introduction of antigenic mutations ( mutation limited ) , to a regime determined by the availability of antigenic niches ( selection limited ) , and under varying strengths of competition between strains ., We use this framework to determine the conditions under which a limited antigenic repertoire could explain the observed phylodynamic patterns of H3N2 influenza ., Explicit modeling of evolution , through the introduction of antigenic mutation at different rates , allows us to consider phylogenetic trees in addition to epidemiological dynamics ., Resulting phylodynamic patterns range from successive strain turnover , to discordant antigenic sets , to dynamics resembling those of a model lacking explicit evolution and finally to the collapse of antigenic structure ., Each of these can be explained by the interplay of selection and mutation , as measured here through the MKR index , and by considering different strengths of immunity generating competition between strains ., The dynamics of our individual-based model are generally in good agreement with the epidemic behavior of influenza A ( Figure 6 ) ., Like observed epidemiological patterns 17 , 23 , 37 , 38 , 39 , 40 , 41 , annual temperate climate epidemics occur almost regularly with substantial year-to-year variation in incidence ( CoV\u200a=\u200a1 . 1±0 . 1 compared to ( CoV\u200a=\u200a1 . 0±0 . 2 ) in literature survey ., Observed temperate climate annual attack rates of influenza A ( H3N2 ) are slightly higher , approximately 8% from 1976 to 1981 38 compared to 5 . 8%±0 . 1 in simulation , while peak epidemic weakly cases are higher in the simulation ( IQR≈200–800 cases per 100000 ) in comparison to ( IQR≈130–380 , IQR≈80–240 ) in 37 and 40 respectively ., The tropics exhibit lower and weaker seasonality ( Figure 6 , Figure S3 ) with slightly lower yearly attack rates ( 5 . 5%±0 . 1 ) and substantially lower prevalence ( Figure 6 ) ., In agreement with antigenic cartography 18 , 42 13±6 clusters dominate the global world population ( Figure S4-A ) with an average duration of 4±2 years , exhibiting mostly the dominance of 1–2 clusters globally ., With respect to individual epitope changes we find the model reproduces the observation of the tropics being “antigenically ahead” 23 , giving rise to antigenic changes 2 . 5±1 . 5 month ahead of the northern and southern hemisphere ( Figure S4-B ) and showing decline in antigenic variants 1 . 7±0 . 2 month earlier than the northern hemisphere ., In agreement with observed phylodynamic patterns 43 the tropics metapopulation has a higher proportion in establishing the trunk ( 68%±9 ) of the phylogeny followed by the northern ( 20%±8 ) and the southern ( 12%±3 ) population ., The higher contribution of East and South-East Asia as the origin of H3N2 globally circulating lineages is hypothesized to originate from lower and mixed seasonality in these regions and is consistent with our model 23 ., The key difference between the hemispheres in the model being , lower population size in the southern hemisphere with proportionally lower contact rate between the meta-populations ., Refinement of the epidemiological model , such as the inclusion of an exposed period , can further improve the comparison to empirical data ., In particular , the above properties were obtained with a basic reproduction number of R0≈3 . 24 , on the upper bounds of current estimates for seasonal influenza ., This value can possibly be decreased by considering such an extension ., We find that a model with 4 epitopes and a low but variable number of variants per site , an antigenic mutation rate of ≈10−5 per day and a reduction of cross-immunity of 13% per epitope results in phylodynamic patterns broadly consistent with those seen in H3N2 influenza ( Figure 6 ) ., When doubling the number of epitopes , the model maintained similar results with respect to genetic diversity π and overall incidence , when the total mutation rate across epitopes was maintained and the cross-immunity was modified to a 6 . 5% per epitope change ., These parameters are quite comparable to parameters used in other models of influenza evolution ., For example , Koelle et al . 18 use 5 epitopes with mutations of either large or small antigenic effect ., Small mutations reduce cross-immunity by 7% and occur at a rate of ∼5×10−4 per day , while large mutations reduce cross-immunity by 20% and occur at a rate of ∼10−5 per day ., In the model of Bedford et al . 21 mutations reduce cross-immunity by between 1% and 11% ( 95% bounds ) , but occur at a faster rate of 10−4 per day ., Ferguson et al . 17 find that a model with 12 codons , each with 20 amino acid variants , in which mutations occur at a rate of 3×10−5 per day and reduce cross-immunity by ∼7% gives restricted diversity without short-term strain-transcending immunity , and 1 . 2×10−4 per day , when transient immunity is included ., From this , it seems clear that models involving a slow influx of antigenic mutants of around 10−5 per day are generally capable of producing influenza-like patterns of restricted diversity ., Increasing host population size in the model results in an increase in viral genetic diversity , as more opportunities for antigenic mutation arise within the larger host population ., Thus , scaling competitive interactions between strains , and/or antigenic mutation rate , is required to maintain limitations on the effective exploration of antigenic space ., In addition , other epidemiological phenomena , besides low antigenic mutation rates , may also contribute to limit the rate at which novel antigenic phenotypes emerge in the influenza population ., These may be provided by population structure and the seasonality of transmission 23 , 44 , as well as by short-term strain-transcending immunity , which was found capable of limiting genetic and antigenic diversity in a similar model with a much larger antigenic space 16 and in a limited diversity antigenic model 45 ., However , a global metapopulation structure is not expected to be the dominant cause behind the low standing genetic diversity of influenza ., Influenza B exhibits similar epidemiological dynamics , and lower prevalence , yet it exhibits much higher genetic diversity through the co-circulation of multiple lineages 16 , 46 ., Also , a more complex metapopulation structure with multiple patches can either increase genetic diversity by facilitating the coexistence of viruses at different weakly coupled patches , or decrease genetic diversity through the generation of population bottlenecks ., The role of variation in viral fitness is an important consideration in future studies , particularly in light of recent observations linking binding properties of HA with antigenic escape 47 ., The empirical finding of a non-trivial relationship between virus fitness in susceptible individuals and immune evasion was suggested as a possible alternative mechanism for generating positive selection pressure on antigenic sites and for limiting antigenic diversity 47 ., Future work should investigate quantitative patterns and statistical approaches for discriminating among the different models and associated hypotheses that currently exist in the literature and for inferring the relative importance of the mechanisms they represent , keeping in mind that the models are not necessarily mutually exclusive ., At the same time , empirical advances on the molecular basis of immune evasion and recognition , on the genotype-to-phenotype map , and on epitope identification and population serology , will allow a better evaluation of the models assumptions , including the representation of serological space ., In common with the Recker et al . 22 model and in contrast with other phylodynamic models 16 , 17 , 21 , we find here that antigenic epitopes are frequently recycled ( Figure S6 ) ., Importantly , this does not mean that such recycling is observed for the antigenic types ( epitope repertoires ) themselves , since the same antigenic type only re-emerges at long intervals ( Figure S5 ) and rarely in the course of 40 simulated years ., Its possible that such reemergence could explain the antigenic cross-reactivity between sera from around the 1918 H1N1 pandemic and viruses emerging in the 2009 H1N1 pandemic 48 , 49 , 50 , 51 ., However , antigenic stasis of the swine lineage leading to the 2009 pandemic could also explain these observations ., Much further work on epitope identification and population-wide serological surveys is necessary to establish the validity of this models prediction on the re-cycling of constituent low diversity epitope variants ( Figure S6 ) ., Nevertheless , several empirical observations are becoming available that are consistent with such recycling and the subject is discussed in detail in the companion paper 52 ., For example , an antigenic analyses performed on H2N2 influenza , a number of monoclonal antibodies raised against a 1957 strain were shown to cross-react strongly with a strain isolated in 1964 , yet not with the 1963 strain 53 ., In Reichert et al . 54 the hemagglutinin of both novel pandemic H1N1 and pre-1940 H1N1 lack specific glycosylation sites on the globular head of HA1 ., These reverse glycosylation patterns were suggested to possibly shield antigenic sites for a timescale of decades and in so doing , to effectively contribute to their recycling and to the age distribution of cases ., In Bui et al . 55 , several protective antigenic and T cell H3 epitopes show temporal variability across drift variants , with two of these specifically exhibiting a decrease and increase in conservancy consistent with epitope “recycling” ., Post translational and conformational changes may hinder the validity of this analysis especially for epitope 1 which acquired two surrounding glycosylation sites ., In Wang et al . 56 , mice Anti-H3 mAbs were shown to neutralize H3 viruses that span 40 years , as measured by immunofluorescence against MDCK cells ( see Table 2 in 56 ) ., All three mAbs ( see Figure 4 in 56 ) displayed variability in their ability to neutralize H3 viruses for lower concentrations ( <15 µg/ml of 7A7 and <25 µg/ml for the other two ) in plaque reduction assays ., For example for mAb 7A7 neutralization was better for HK68 , than diminished for BJ92 and then increases for PAN99 and BRIS07 ., This pattern could also be due to secondary effects of amino acid differences outside the actual epitope as well through structural effects , but effectively behaves as epitope recycling over substantial durations of many years ., In conclusion , within our framework , the rate of antigenic mutation was found to strongly influence whether selection was positive or negative , and hence , the topology of the tree and associated diversity of the virus ., Strong positive selection is generated by effective competition under low mutation rates , and results in spindly trees with low genetic diversity ., In this regime , antigenic mutations often fix in the virus population , lowering genetic diversity , as consistent with H3N2 ., An increase in mutation rate across a broad spectrum in competition strength , leads to negative selection and generates antigenic divergence ., This can potentially result in the coexistence of discordant antigenic types repressing the emergence of antigenic hybrids , through strong negative selection on antigenic change , with each discordant antigenic type maintaining a deep phylogenetic branch ., Although not strictly mirroring the assumptions about development of the Recker et al model , our framework strongly implies that limitations on antigenic architecture alone are unlikely to reliably reproduce “skinny trees and some restrictions on mutation rate and/or other considerations such as fitness differences are likely to play a role ., It is important to note that this exercise does not also privilege other hypotheses concerning diversity restriction in influenza as these also are strongly sensitive to mutation rate ., Overall , it emphasizes that phylogenetic patterns do not serve as a discriminatory tool between these by no means mutually exclusive hypotheses ., However , they can provide a basis to exclude specific hypotheses and offer a means by which the contributions of mutation and selection can be assessed ., Needless to say , the latter has important implications for the updating of vaccines against influenza ., Under a mutation-limited regime , a hypoth

Introduction, Results, Discussion, Materials and Methods

Influenza A ( H3N2 ) offers a well-studied , yet not fully understood , disease in terms of the interactions between pathogen population dynamics , epidemiology and genetics ., A major open question is why the virus population is globally dominated by a single and very recently diverged ( 2–8 years ) lineage ., Classically , this has been modeled by limiting the generation of new successful antigenic variants , such that only a small subset of progeny acquire the necessary mutations to evade host immunity ., An alternative approach was recently suggested by Recker et al . in which a limited number of antigenic variants are continuously generated , but most of these are suppressed by pre-existing host population immunity ., Here we develop a framework spanning the regimes described above to explore the impact of rates of mutation and levels of competition on phylodynamic patterns ., We find that the evolutionary dynamics of the subtype H3N2 influenza is most easily generated within this framework when it is mutation limited as well as being under strong immune selection at a number of epitope regions of limited diversity .

Influenza A ( H3N2 ) has circulated in the human population since 1968 causing considerable annual morbidity and mortality worldwide ., Despite the rapid evolution of the hemagglutinin ( HA ) protein and strong diversifying selection , the global virus population is characterized by a low standing diversity , evident in the serial replacement of antigenic types and in the ‘cactus-like’ structure of its genealogical tree ., Elucidating the mechanisms behind these puzzling patterns is key to understanding the evolution of seasonal ( H3N2 ) influenza ., One recent epidemiological model proposes a restricted set of antigenic types whose waves of dominance result from frequency-dependent immune selection ., Here we develop a model of limited antigenic diversity that explicitly incorporates mutational processes , and use it to address , first , whether this type of antigenic space is capable of generating the characteristic phylogeny of HA sequences , and second , whether the dynamics of ( H3N2 ) influenza are primarily limited by the arrival of mutations or by the opening of antigenic niches ., We conclude that a limited antigenic space can explain the observed phylogenetic patterns and that a limited mutation rate is a key property underlying the dynamics of ( H3N2 ) influenza ., Our study provides a general framework for assessing the relative roles of selection and mutation in a variety of infectious disease systems .

public health and epidemiology, evolutionary ecology, ecology, evolutionary biology, virology, epidemiology, infectious disease epidemiology, biology, computational biology, microbiology, population biology, viral evolution

journal.pcbi.1004365

Neutral Models of Microbiome Evolution

Our framework applies to a population of hosts and an available pool of microbial colonists ., As a first step , we assume that hosts do not exert any preferences on the microbial taxa they acquire ., Similarly , we assume that microbes do not interfere with host reproductive capacity , or the survivorship and reproductive success of other microbes in the community ., As will become clear , the only indirect effects that influence microbial recruitment and persistence from one generation of hosts to the next are competition for space within hosts and the relative abundance of microbial taxa ., Simply put , we assume that the ecological and evolutionary processes that operate on hosts and their microbiomes are neutral; in this regard , our framework is analogous to neutral theories in evolutionary biology 36 , 37 , ecology and biodiversity 34 ., We expand on this analogy later , but for now , we note that neutral theories provide parsimonious accounts of the types of patterns that can emerge in complex systems , they serve as null models for statistical hypothesis tests , and they provide platforms upon which we may construct more elaborate representations of these same systems 38 ., In this framework , hosts reproduce asexually in discrete generations , following a neutral Wright-Fisher process 39 , 40 , where each individual in a succeeding generation chooses a parent randomly from the preceding generation ., Hence , with a population of hosts of constant size N , all asexual individuals will share a common ancestor after 2N generations , on average ., In our models , asexual reproduction is a computational convenience , and can be replaced with sexual reproduction without changing the essential patterns that we observe ., We model how hosts acquire their microbiomes in three ways ( Fig 1 ) ., First , under a strict “parental-acquisition” ( PA ) process , all hosts acquire their microbial communities directly from their parents ., Second , with strict “environmental-acquisition” ( EA ) , hosts acquire their microbiomes solely from the environment ., Between these two extremes , we also allow a third “mixed-acquisition” ( MAx ) process , whereby hosts acquire some percentage , x% , of their microbiomes from their parents and ( 100-x ) % from the environment ., MA0 is exactly equivalent to EA , and MA100 to PA; as such , EA and PA designate boundary conditions of the ecological processes that mediate microbial acquisition in hosts ., It is worth pausing at this point to clarify what we mean when we say that hosts acquire their microbiomes from their “parents” or their “environments” ., Our models do not explicitly take account of the life events–illness , infections , changes in environments or diets–of each host within a generation , nor does it consider microbial fluxes within the lifespan of each host ., Instead , the microbial composition of each host is essentially measured as an aggregate over the single generation that the host exists ., Consequently , when we quantify the percentage of microbes from parents and environment using , say , MA10 , we mean that over the life of the host , 90% of its microbes come from the environment and 10% from its parent ., In our models , it is possible that the parental contribution happened in the first 10% of the host’s life , or it may be that over the entire lifespan of the host , there was an ongoing contribution by the parent that amounted to 10% of the microbial composition ., Since we allow hosts to recruit microbes from an “environment” , we need to define how the microbial content of this environment is constituted ., In simulations , we characterize microbial composition using a distribution of taxa’ relative abundances ., We propose three processes that determine the composition of the pool of microbes available for recruitment ., First , we assume that the environment has a microbial composition that remains fixed over time ., For the “fixed environment” ( FE ) , all taxa are present in the environment throughout the simulation , and are available to every generation of hosts ., The second process we propose involves a changing environmental microbial profile , whereby the relative abundance of each microbial taxon available to the hosts in a given generation , is an aggregate of their abundances from all hosts of the preceding generation ., Under this “pooled-environment” ( PE ) , microbial composition is reflection of what was present in the parents of the current generation of hosts ., A third , intermediate , process is a combination of the previous two “environments”: the environmental microbial pool available for recruitment contains a percentage , y% , from the parental pool of microbes , and ( 100-y ) % of microbes from the fixed environment ., Under this “mixed environment” ( MEy ) , the proportion of contribution from host microbiomes is given by y ., As with our acquisition models , ME0 and ME100 are equivalent to the boundaries FE and PE , respectively ., Our framework allows us to combine different host-acquisition processes with different ways of constructing the pool of available microbes in the environment ., Conceptually , each of these combinations is a particular neutral model , capturing some of the elements previously discussed in the literature ., For instance , PA or MAx incorporate the phylogenetic dependencies that Yeoman et al 31 discuss , and EA x PE is equivalent to what Costello et al 30 call dispersal limitation , whereby the local host community influences microbial composition ., It is worth noting that the combinations PA x ( FE , MEy , PE ) –read as “PA in combination with FE , with MEy or with PE”–will give identical results ., This is because , in all cases , the environment contributes nothing to host microbial content ( see the first row in Fig 1 ) ., We model the construction of the microbial community in each host by competitive random sampling with replacement ., Under this process , each host allows only a fixed and limited number of microbes to populate its microbiome ., If microbial acquisition occurs under EA , each host samples randomly from the available pool of taxa according to the relative abundance of each taxon in the environment ., In the case of MAx , x% of microbes are selected from the parent and ( 100-x ) % from the environment ., If hosts acquire their microbial taxa under PA , then all microbes are inherited from the hosts’ parents , although the relative abundance of each taxon fluctuates multinomially ., By constructing microbial communities in this way , we allow stochastic factors and indirect competition to modify taxon composition within and between hosts , as proposed by Costello et al 30 ., By simulating combinations of PA , MAx and EA against FE , MEy and PE forward in time over many host generations and over a range of conditions , we are able to recover data on the behavior of individual microbial taxa , as well as a variety of summary statistics , including the expected time it takes individual taxa to invade all hosts or go extinct in the host population , and the trajectories of microbial taxonomic richness ( measured simply as the total number of microbial taxa ) and microbial taxonomic evenness ( measuring the similarity in the frequency of each taxon ) , microbial diversity within hosts ( α-diversity ) , inter-host variation in microbial composition ( β-diversity ) and the aggregate microbial diversity from all hosts in the population ( γ-diversity ) ., Here , we report only on the latter three measures of diversity ., Microbial diversity within the host population is a function of the proportion of microbes that parents contribute directly to offspring and the proportion they contribute to the environment ., Fig 2 illustrates how population-level taxon abundances change under various combinations of these proportions ., In our simulations , the distributions of taxon abundances under high levels of parental contributions are skewed , and may be approximated by commonly-applied distributions , including the log-normal distribution and the Dirichlet multinomial ( DM ) distribution 41 ( Fig 3; the DM distribution has the advantage of allowing α- , β- and γ-diversities to be simulated–see S2 Fig ) ., The ability to recover skewed abundance distributions is interesting , because we begin our simulations with a uniform distribution of microbial taxa , and we retain this uniform distribution in the fixed environment throughout the evolutionary history of the host population ., Consequently , the emergence of dominant and rare taxa is a consequence of repeated parental contributions either directly to the next generation of hosts or indirectly to the environment ., In fact , all simulations in which there was complete parental acquisition of microbes ( i . e . , PA ) resulted in the loss of all but one microbial taxon in the host population ., Similarly , when the environment was reconstituted each generation exclusively with microbes from the parents ( i . e . , PE ) , the same pattern was observed with only a single microbial taxon remaining ., These result are consistent with predictions made under neutral models of community ecology 42 , and highlight the strong depressive effect of parental transmission , either directly from parent to offspring or via parental contributions to a local pool of microbes , on population-level microbiome diversity ., With EA x FE , microbes are obtained randomly from a fixed environment that persists over the evolutionary history of the hosts; unsurprisingly , the host population retains all microbes found in the environment ., Interestingly , when microbes are obtained both from parents and a fixed environment ( MA x FE ) , we still see the persistence of all or almost all microbes in the host population ( see Fig 2A and 2B , first column of each bar chart; ME ( 0 ) is equivalent to a fixed environment with no microbial contributions from parents ) ., This is true even when the proportion of microbial taxa that an individual host acquires from the fixed environment at each generation is very small , on the order of 0 . 001 ., Therefore , a very small contribution from a constant environmental source of microbes is sufficient to retain high levels of microbial diversity in the host population ., Microbial diversities are frequently measured in three ways: α-diversity , β-diversity , and γ-diversity ., Our simulations indicate that all three measures depend on the percentage of parental contribution to offspring microbiomes and the composition of the environmental microbial pool ( see S1–S3 Tables for simulation means and standard deviations ) ., Under our neutral model , in which the absence of host sub-population structure means that all hosts sample their microbes from the same environment , α- and γ-diversities remain high , and β-diversity remains low , for a large part of the range of direct or indirect parental contributions ( i . e . , to offspring or to the environment , respectively ) ., Nonetheless , at high values of parental contributions , there are discernible differences in diversities , and we have also focused our simulations in these areas ( Fig 4; see S4–S6 Tables for simulation means and standard deviations ) ., In general , α-diversity ( average diversity within hosts ) and γ-diversity ( overall diversity within the entire population of hosts ) increase as we increase the fixed environmental contribution because a fixed environment helps maintain a uniform distribution of taxon abundances and delays the loss of microbial taxa during evolution ., Conversely , when hosts acquire increasing proportions of their microbiomes from their parents directly , or indirectly from a pooled environment , the variation of taxon abundance increases and taxon richness tends to decrease , thus lowering both α- and γ-diversities ( Fig 4A and 4B; S1 , S2 , S4 and S5 Tables ) ., Inter-host variation in microbial composition , or β-diversity , also depends on the degree of parental inheritance , and the ratio of fixed-to-pooled environmental components ( Fig 4C; S3 and S6 Tables ) ., Under the combination of PA x ( FE , ME or PE ) , β-diversity tends to zero , because all hosts descend from a single common ancestor and , as noted above , only a single microbial taxon remains in all hosts ., When we have the parental microbiome as the only source of microbiomes in the next generation , ultimately , all lineages will have acquired their microbiomes from the most recent common ancestor ( MRCA ) of the population of hosts ., Additionally , from one generation to the next , stochastic sampling of microbes over evolutionary time will result in the loss of all but one microbial taxon ., Interestingly , with a high percentage of environmental acquisition , β-diversity is also relatively low , because all hosts acquire a large proportion of their microbial taxa from the same environmental pool , and consequently , will tend to acquire the same set of taxa ., As noted above , the highest β-diversity occurs in a relatively narrow range of values of pure parental acquisition ( between 87–99% of direct parental transmission; S6 Table ) ., If we focus on the relationship between β-diversity and the environmental pool , we see that its behavior is similar to that of α- and γ-diversities: it decreases as we increase the pooled environmental contribution to offspring microbiome ., This is because a pooled environment , with contributions from the parental generation , tends to give rise to a non-uniform distribution of microbial taxa ., As the degree of parental contribution increases , the environmental community will be dominated by few highly abundant species which are likely shared by most or all hosts within the population , accounting for high between-host similiarity in microbial composition ( Fig 4C; S3 and S6 Tables ) ., It is important to note that our simulations have not been performed with inference or prediction in mind: the number of hosts , the number of microbes , and the number of taxa in our simulations are not necessarily equivalent to those of real-world microbial communities and their hosts , nor have we necessarily chosen the appropriate diversity indices or taxonomic resolution to optimize prediction/inference ., Nonetheless , it is helpful to examine how the simulated values of diversity compare to empirical observations , and what these comparisons might tell us about the evolutionary processes that are acting on microbiomes ., As an example , we used genus-level taxonomic data from the NIH Human Microbiome Project ( HMP ) 43 , specifically , a table of relative abundance found in different compartments of the human body ( http://www . hmpdacc . org/HMSMCP/; see S1 Dataset ) ., Several large samples from the anterior nares , vaginal posterior fornix , stool , buccal mucosa , tongue dorsum and supragingival plaque were chosen to calculate α- , β- and γ-diversities on genus level ( Table 1 ) ., Values of α- and γ-diversity obtained from all sampled sites of the human microbiome are low , in comparison to most of the values we obtained in our simulations ., In fact , human microbiome diversities are generally lower than those of other non-human primates 44 , 45 ., If we compare the empirical diversities to those obtained in our simulations ( S1–S6 Tables ) , we would have to posit very high parental contributions , both direct and indirect ( >90% ) , to account for the α- and γ-diversities across all human body sites ., In contrast , values of β-diversity appear to provide a little more discrimination amongst body sites: the site with the lowest β-diversity is the vaginal posterior fornix , and its value is consistent with a very low degree of direct parental contribution in our simulations ( approximately between 0–15% ) ., The β-diversities at other sites appear to suggest higher levels of parental contribution ( again , >90% ) ., In the next section , we discuss the implications of these results , as they relate to human microbiome evolution and how the neutral model may be used to construct hypotheses about relevant evolutionary and ecological processes ., In this paper , we introduce a simple and flexible framework to model the evolution of microbiomes within a population of hosts , which takes account of different modes of microbiome acquisition and environmental microbial composition ., Under our neutral model , microbiome composition is affected by sampling effects ., Stochastic changes in microbial abundances may affect the persistence of microbial taxa in the microbiome over one or a few generations ( i . e . , ecological drift ) , or over many generations ., The latter may occur because host lineages die out; when this happens , changes in microbial abundance across the whole population of hosts are essentially equivalent to changes in allele frequencies ( i . e . , genetic drift ) ., The constitution of the microbial community in the environment also plays a considerable role in determining the ultimate fate of microbial taxa within host microbiomes ., With a fixed environment , when there is a constant pool of the same microbial taxa from one generation of hosts to the next , microbial taxa never go extinct from the host population as long as hosts obtain some fraction of their microbiome from the environment ., This is true , even when that fraction that the environment contributes to each host’s microbiome is very small ( e . g . , 0 . 1% ) ., In contrast , when the environmental composition of microbes reflects the microbial content of the hosts in previous generations ( i . e . , PE , the “pooled” environment in our model ) , microbial diversity of the environment shrinks , as does the diversity of host microbiomes ., Therefore , the extent to which parents contribute to the microbiomes of their offspring ( either directly or through their contributions to the pooled environment ) plays a crucial role in shaping microbiome diversity and constitution ., In our simulations , values of α- and γ- diversities are at their lowest when parental contribution to the microbiome is high ., Inevitably , microbial taxa are lost from the population as host lineages are lost ., Thus , under our neutral model , it is possible to recover skewed microbial abundance distributions reminiscent of those obtained with real data , despite a fixed environmental component that remains uniform and constant throughout our simulations ., Increasing skewness–essentially , decreasing eveness–is obtained as we increase the degree of parental inheritance ., Of course , we don’t claim any deep insight here: no one should be surprised that we are able to recover skewed abundance distributions with our models , because there is a large body of literature on the mechanisms–both neutral or otherwise–that may lead to the emergence of skewed abundance distributions ( see 46 for an excellent synthesis ) ., Our results reinforce what others 34 , 47 , 48 have found , by adding yet another neutral mechanism to account for the emergence of skewed abundance distributions ., Our framework includes sampling effects on an undivided host population , which evolves under a Wright-Fisher process ., Consequently , our models have some points of similarity with those that have been developed in population genetics ., For instance , Orive et al 49 analyze the evolutionary dynamics of endosymbionts using a discrete-time Moran population genetic model ., In their model , endosymbionts are acquired either vertically , passed on from parent to offspring , or horizontally from the environment ., This corresponds to our MA x FE model and , in agreement with our results , Orive et al find that increasing the environmental contribution of endosymbionts to host cells results in greater diversity within cells and less diversity between cells ., Our models do not include any mutational process or speciation acting on the microbes , as time moves forward ., In reality , of course , microbes acquire mutations in their genomes at a rapid rate , but the measures of diversity we use in our analyses capture differences in taxonomic composition , not genetic diversity ., In our models , it is implied that no cladogenetic events have occurred over the course of the simulations ., The models presented here provide an opportunity to construct hypotheses , and make qualitative predictions , about the patterns of diversity we can expect to find in different biological situations ., For example , the effects of “pure” pooled versus fixed environments on microbiomes can be found in a comparison of social and solitary bees ., Social bees exhibit behaviors that are likely to result in the transmission of microbes from a microbial pool within the colony 50 ., In contrast , solitary bees acquire their microbiomes from the environment , through feeding or burrowing ., Our model would predict that social bees would have lower α-diversity and lower taxonomic richness than solitary bees ., This is consistent with the results obtained by Martinson et al 51 who surveyed the microbiomes of eusocial bee species Apis spp ., and Bombus spp ., and non-social bees ( 11 species ) and wasps ( 3 species ) : they found depauperate microbiomes in social bees compared to non-social bees ., Whereas it is reassuring to obtain empirical corroboration for our models , arguably neutral models are most useful when real-world observations run counter to predicted outcomes ., Falsification of neutral models provides a justification for augmenting these models to include additional processes that account for the phenomena under study ., In this regard , our analysis of the Human Microbiome Project data is instructive ., As we have noted above , our simulations should not be used for inference , and we should be cautious about reading too much into the comparisons between empirical and simulated patterns of diversity ., Nonetheless , at least for some sites , i . e . , the stool , the tongue dorsum , the supragingival plaque , the anterior nares and the buccal mucosa , the low empirical values of α- and γ- diversities appear to point consistently to a high level of parental inheritance when compared against values obtained in our simulations ., There is evidence that the human microbiomes at various sites are seeded at birth by the mother , particularly if this birth is through the vaginal tract 52 ., There is also reasonably strong evidence that families share microbes to a greater extent than unrelated individuals in a population 53 , and at least in some human populations , mothers share more microbes in common with their offspring than with unrelated children 54 ., It is not clear , based on the studies that have been done to date 55 , whether the values of direct or indirect parental contribution we obtain when we compare empirical and simulated diversities are significantly higher than would be obtained in real populations , but we expect that the intuition of mosts microbial ecologists is that percentages of direct and pooled parental contributions > 90% are likely to be too high ., Putting to one side the caveats about inference , we accept that while this intuition does not constitute evidence against the neutral model , it is likely to engender scepticism about the model’s correctness ., If it is , in fact , true that direct or indirect parental contributions to the next generation’s microbiomes are not as high as our simulations suggest , how do we account for the apparent depression in α- and γ-diversities , and elevation of β-diversities at these sites ?, One hypothesis that explains low α- and γ-diversities , and high β-diversities , and does not require the action of non-neutral processes , is the existence of local host subpopulations ., The existence of subpopulations of hosts , with limited immigration and sharing of microbes between subpopulations , is likely to give the appearance of high parental contribution from one generation to the next ., Certainly , this is a plausible explanation for patterns of microbiome diversity in the oral cavity ( i . e . , the buccal mucosa , tongue dorsum and supragingival plaque ) and stool samples , because of the likely influence of familial 53 or cultural dietary preferences/practices 56 or lifestyles 57 on these microbiomes ., A similarly explanation may account for patterns of microbiome diversity of the anterior nares ., The vaginal posterior fornix presents an interesting contrast to the other body sites because the α- and γ-diversities suggest high parental contributions ( although they cannot distinguish between direct or indirect contributions ) , whereas β-diversity suggests a low direct parental contribution ., This inconsistency may again cause us to reject the neutral model in favor of an alternative explanation , but in this case , subpopulation structure may play a minor role relative to selection for a vaginal microbial community that is common amongst hosts ., Such a selective filter is likely a consequence of a complex suite of factors including host immune defences , hormonal cycling , pregnancy , and the presence of apparently beneficial microbial species ( e . g . , Lactobacillus spp . ) 58 ., This hypothesis explains both the high level of α- and γ-diversity ( i . e . , a few abundant species with many rare species ) , and the low β-diversity ., For the human microbiome , neutral models have the potential to help identify additional processes that may account for patterns of diversity ., As noted , of the two processes identified above–host subpopulations and selective filters–the former still remains part of an underlying neutral process , and a plausible extension to the neutral framework presented here ., Rejection of a simple neutral model therefore allows us to identify incremental additions that may increase explanatory power ., Another example of empirical data that appears to contradict the expectations of our models is the comparison of microbiome diversities in high microbial abundance ( HMA ) and low microbial abundance ( LMA ) sponges 59 ., HMA sponges have large numbers of associated microbes , in contrast to LMA sponges ., Additionally , researchers have shown that microbial diversity in LMA sponges is lower than that of HMA sponges 60 , 61 ., Based on our results , we would predict that there is a greater degree of vertical transmission in LMA sponges , but it turns out that this is not the case: Schmitt et al 62 have found that “vertical transmission , as a mechanism to obtain bacteria , seems to occur mainly in HMA sponges” ., Giles et al 60 propose two possible reasons to account for the low diversity in LMA sponges ., First , there may be selective filters that permit only certain microbial taxa to colonize the sponges; second , the initial colonization event is stochastic , but serves to constrain or exclude successive colonizations ., As with the human microbiome data , the sponge example is important because it does not rely a priori on non-neutral processes to account for the low diversity in LMA sponges; instead , selection ( or other ecological and/or evolutionary processes ) is invoked only after it is shown that vertical transmission in LMA sponges is unlikely , thus indicating that our neutral models are an inadequate explanation for the observed data ., Riffing on the theme that “Essentially , all models are wrong , but some models are useful” 63 , Hubbell , writing about models in community ecology , says “Probably no ecologist in the world with even a modicum of field experience would seriously question the existence of niche differences among competing species on the same trophic level” 64 ., But , he continues , “Neutral theory begins with the simplest possible hypothesis one can think of … and then adds complexity back into the theory only as absolutely required to obtain satisfactory agreement with the data” ., We agree with Hubbell: to paraphrase , given what we know about the interplay between hosts , their microbial communities , and the environment , we would hesitate to put money on the table and bet that many microbiomes have evolved under the simplest neutral models that we have constructed here ., But we would be equally hesitant betting in favor of the null hypotheses evaluated in statistical tests of significance ., The value of these hypotheses resides not in their rightness or wrongness but in their ability to protect against overconfidence in our favorite , more complex model ., Whereas it is true that biological processes are frequently complex , Occam’s Razor dictates that we construct as simple explanations ( or models ) as possible ., In this way , we remain vigilant against the addition of unnecessary and unjustifiable complexity ., Much as we do with statistical hypothesis tests , we accept stronger alternative explanations only when we are sufficiently confident that our neutral hypotheses are unlikely ., This is not to say that neutral models only serve as strawmen; in molecular evolution , for instance , neutral models are frequently effective at explaining molecular variation 65 ., And even in cases when the assumption of neutrality is questionable , the use of neutral models of substitution applied in molecular phylogenetics does not appear to jeopardize the accuracy of tree reconstruction 66 ., Consequently , without taking account of the evolutionary processes of mutation , speciation , selection or recombination , or the ecological processes that operate in the context of spatial , environmental , and temporal heterogeneity , what we have developed is a framework on which we can begin to evaluate empirical patterns of diversity , and where necessary , add more elaborate ecological and evolutionary scenarios ., We believe that even this simple framework , devoid as it is of all the embellishments afforded by evolution and ecology , can serve a useful purpose: it is a suitable staging ground on which we can construct null models of microbiome diversity in populations of hosts and it allows us to make strong , testable predictions ., Simulated host populations consisted of a fixed number of virtual host individuals ( N = 500 ) ., Each host was allocated a virtual microbiome with a limited capacity or slots of microbes ( n = 1000 ) ., The environmental pool consisted of 150 microbial taxa ., Large number of hosts ( N = 2000 ) , microbes ( n = 100000 ) per host and microbial taxa ( m = 500 ) were also simulated with our neutral model , and similar patterns of diversity were observed ( S1 Fig ) ., The microbiomes of the initial generation of host individuals were seeded randomly , with bacteria sampled from a uniform distribution of taxon abundances ., We used an initial uniform distribution of taxa because we wanted to ascertain whether the equilibrium distribution of abundances obtained at the conclusion of our simulations would recover patterns seen in natural microbiomes ., For each subsequent generation , the microbiome of each individual host was simulated by populating each of the available slots in the individual\s microbiome by sampling microbial taxa with replacement ( multinomial choice ) from either the environment ( with probability given ( 1-x ) ) or from the microbiome of a parent host individual ( selected with uniform random probability from the population of the previous generation ) ., When sampling from a parental/environmental microbial community , the probability that the new host microbiome will acquire a particular microbial taxon is given by the relative abundance of that taxon within the community ( see below for details on how environmental microbial taxon abundances were calculated ) ., The probability , x , that a particular slot in a new individual host\s microbiome was occupied by a microbial taxon sampled from a randomly selected parent was varied across simulations ., Two sets of simulations were performed: ( 1 ) x and y varied linearly , between 0 and 1 , with increments of 0 . 1 ( see S1–S3 Tables for means and standard deviations of diversities ) ; and ( 2 ) with values of x , y ∈ ( 0 . 0 , 0 . 50 , 1 , 2…10 ) ( see S4–S6 Tables for diversities ) ., When x = 0 . 0 , a host’s microbiome was sampled

Introduction, Results, Discussion, Materials and Methods

There has been an explosion of research on host-associated microbial communities ( i . e . , microbiomes ) ., Much of this research has focused on surveys of microbial diversities across a variety of host species , including humans , with a view to understanding how these microbiomes are distributed across space and time , and how they correlate with host health , disease , phenotype , physiology and ecology ., Fewer studies have focused on how these microbiomes may have evolved ., In this paper , we develop an agent-based framework to study the dynamics of microbiome evolution ., Our framework incorporates neutral models of how hosts acquire their microbiomes , and how the environmental microbial community that is available to the hosts is assembled ., Most importantly , our framework also incorporates a Wright-Fisher genealogical model of hosts , so that the dynamics of microbiome evolution is studied on an evolutionary timescale ., Our results indicate that the extent of parental contribution to microbial availability from one generation to the next significantly impacts the diversity of microbiomes: the greater the parental contribution , the less diverse the microbiomes ., In contrast , even when there is only a very small contribution from a constant environmental pool , microbial communities can remain highly diverse ., Finally , we show that our models may be used to construct hypotheses about the types of processes that operate to assemble microbiomes over evolutionary time .

Microbial communities associated with animals and plants ( i . e . , microbiomes ) are implicated in the day-to-day functioning of their hosts ., However , we do not yet know how these host-microbiome associations evolve ., In this paper , we develop a computational framework for modelling the evolution of microbiomes ., The models we use are neutral , and assume that microbes have no effect on the reproductive success of the hosts ., Therefore , the patterns of microbiome diversity that we obtain in our simulations require a minimal set of assumptions relating to how microbes are acquired and how they are assembled in the environment ., Despite the simplicity of our models , they help us understand the patterns seen in empirical data , and they allow us to build more complex hypotheses of host-microbe dynamics .

journal.pgen.1002222

The Repertoire of ICE in Prokaryotes Underscores the Unity, Diversity, and Ubiquity of Conjugation

Prokaryotes , both bacteria and archaea , have remarkably plastic genomes because they can acquire genetic information at high rates by horizontal transfer from other prokaryotes ., This allows them to adapt rapidly to specific niches and results in large differences in gene repertoires among closely related strains 1–3 ., Three major mechanisms allow gene transfer: natural transformation , transduction and conjugation ., Natural transformation is controlled by the receptor cell and mostly implicated in DNA transfer within species leading to allelic recombination 4 ., Both transduction and conjugation are more invasive , since the recipient has little control over both processes which change gene repertoires dramatically and allow transfer between distant lineages ., Conjugation , in particular , can lead to the transfer of very large fractions of genomes and even entire chromosomes in one single event 5 , 6 ., Several studies suggest that conjugation is the preponderant mechanism of horizontal gene transfer between distant lineages 7 , 8 ., Such cross-clade transfer might be at the origin of the rapid spread of antibiotic resistance through most major lineages of bacterial pathogens in the last few decades 2 , 9 , 10 ., Conjugative elements are also known for encoding other adaptive traits such as toxins , transporters and many secreted proteins including enzymes of industrial interest 11 , 12 ., Conjugation involves a relaxase ( MOB ) , which is the key element in a multiprotein DNA-processing complex , a type IV secretion system ( T4SS ) and a type IV coupling protein ( T4CP ) ( reviewed recently in 13 ) ( Figure 1 ) ., The relaxase binds and nicks the DNA at the origin of transfer ., The relaxase-DNA nucleoprotein complex is then coupled to the T4SS by the T4CP ., The T4SS translocates the relaxase-DNA complex through the membrane of the donor cell delivering it to the cytoplasm of the recipient cell ., The T4SS is a large complex of proteins spanning from the cytoplasm to the extracellular space , including an ubiquitous ATPase ( VirB4 or TraU ) , a set of mating-pair formation ( MPF ) proteins ( from a minimum of 12 to more than 20 ) that elaborate the transport channel , as well as a pilus that allows the attachment to the recipient cell and thereby the translocation of the relaxase-DNA complex ., Protein homology of MPF genes allowed the clustering of all known proteobacterial T4SS into four groups 14 , named after one model of each group , the vir system of the Ti plasmid ( MPFT ) 15 , the F plasmid ( MPFF ) 16 , the R64 IncI plasmid ( MPFI ) 17 and the integrative conjugative element ( ICE ) ICEHIN1056 ( MPFG ) 18 ., For other taxonomic clades , the genes associated with the T4SS , apart from VirB4 , the T4CP and the relaxase , are poorly characterized ., Once the relaxase-DNA complex is in the recipient cell , the T4CP translocates the full DNA and the relaxase ligates the two ends of the DNA into a single circular molecule ., At the final stage of the conjugation process , the element exists in ssDNA state in both cells and the hosts replication machineries are recruited to replicate them to reconstitute the original dsDNA molecules 13 ., A self-transmissible conjugative element must thus comprise three components: the relaxase , the T4CP , and the T4SS ., While most described conjugative systems are located in plasmids , the last decade has seen a growing interest in conjugative systems integrated in chromosomes ( ICEs ) , which include the so-called “conjugative transposons” or “integrated conjugative plasmids” 19 , 20 ., The conjugation of ICEs is poorly documented but is generally assumed to resemble that of plasmids , with a preliminary step of excision with circularization and an additional final step of re-integration in the genome ( Figure 1 ) ., For these steps , some ICEs encode supplementary genes resembling those of temperate phages , e . g . integrases of the lambda tyrosine-recombinase family 21 , 22 , which have led to their classification as “phage-like elements” ., Other ICEs integrate in the chromosome , or excise from it , by using other tyrosine-recombinases 23 , 24 , DDE-transposases 25 , serine-recombinases 26 or by homologous recombination with chromosomal copies of transposable elements 27 , 28 ., Contrary to plasmids , there is little evidence of ICEs replication in cells ( but see , for instance , 29 ) so it is often assumed that they cannot be stably maintained in an extra-chromosomal state 20 ., While ICEs , by definition , are conjugative elements , many other mobile elements populate prokaryotic genomes ., Integrative mobilizable elements ( IMEs ) do not code for a T4SS but can use one coded by other elements just like mobilizable plasmids 30 , 31 ., Genomic islands are integrative elements that can be mobilized by conjugation when they have compatible origins of transfer 32 or by integrating in conjugative elements 33 ., Yet , like for non-mobilizable plasmids , the exact mechanism of mobility of most of these elements remains obscure 34 ., Finally , some chromosomes encode T4SS that are not involved in conjugation but in other processes such as protein secretion and natural transformation 35 , 36 ., It has been suggested that these T4SS probably derived from ancestral conjugative systems 37 ., The presence of an ICE can in principle be assessed by the observation of a conjugative T4SS within a chromosome ., Since it is presently known how to class transmissible plasmids 14 , it should be possible to do the same for ICEs ., There are however important difficulties in this process ., First , it is not known if all ICEs conjugate like plasmids ., The family of conjugative elements of ICEHin1056 was proposed to exist exclusively as ICEs 18 ., Even though a few rare conjugative plasmids of this family were subsequently identified 14 , there might be other families exclusive to ICEs ., Second , the presence in chromosomes of T4SS not used for conjugation may obscure the identification of conjugation systems if no relaxase is present at the locus ., Third , the most conserved proteins involved in conjugation are ATPases ., Finding them in genomes and distinguishing them from other ATPases is challenging ., Fourth , ICEs that are non-functional because of pseudogenization might be difficult to distinguish from functional elements ., In this work we present the results of a scan of prokaryotic genomes for conjugative systems in plasmids and chromosomes and the subsequent analysis to understand their functional and evolutionary relations ., Previous studies provided precious insights of ICE evolution by analyzing closely related ICEs 38 , 39 ., Here we take the complementary approach and aim at the bigger picture ., By searching for conjugative elements in all sequenced chromosomes and plasmids , we quantify the number of ICEs , characterize their diversity in terms of mechanism and phylogenetic representation , and study their evolution at the light of that of conjugative plasmids ., If our assumption that ICEs and plasmids use similar conjugation machineries is correct we should be able to identify ICEs by using the sequence information of a large panel of proteins involved in plasmid conjugation ., Previously , we carried out an analysis of plasmids by performing iterative similarity searches followed by protein clustering 14 , but this approach poses problems of lack of convergence when using chromosomal data ., Profile hidden Markov models ( HMM ) can retrieve more distant similarities than BLAST and do not pose as many problems of convergence as PSI-BLAST 40 ., We therefore built protein profiles of the major representatives of the conjugation machinery using the information on proteins used in plasmid conjugation: relaxases ( MOB types ) , T4CPs and VirB4s ( see Materials and Methods ) ., Additionally , we built profiles for proteins characteristic of each of the 4 types of T4SS found in plasmids of proteobacteria ( see Materials and Methods ) ., By using this approach , we did not need to use ad hoc methods to separate the ATPases ( VirB4 and VirD4 ) because the hits of their profiles did not cross-match significantly ., HMM protein profiles do not use the information of the new hits to change the protein profiles so they can be used reproductively upon change of the databank and independently of any reference dataset ., We will soon make all the protein profiles available to the community by a web server ., All the results of this scan are available in Dataset S1 , including composition of all hits , accession numbers , gene names ( with synonyms ) , and location in the replicons ., We scanned 3 , 489 replicons for the presence of conjugative systems , including 1 , 207 chromosomes , 891 plasmids sequenced along with chromosomes ( PSC ) and 1 , 391 plasmids that were sequenced alone , i . e . without the host chromosome ( s ) ( PSA ) ., Our analysis identified over 7000 proteins with significant matches ( Figure 2 ) ., Close co-occurring hits were clustered together and this allowed the identification of putative T4SS ., When a MOB and a T4CP neighbored a T4SS this locus was regarded as a conjugative system ( see Materials and Methods ) ., Conjugative loci in chromosomes were named ICEs ., Our present results with plasmid sequences were very similar to those previously published 14 ( see Methods ) ., The comparison between chromosomes and the accompanying PSC plasmids allows an unbiased quantitative comparison between plasmids and ICEs in that both sets reflect the same sampling ., Hence , we will show the results on all plasmids only when explicitly mentioned , otherwise all results concern the PSC plasmids ., If we are correct in assuming homology between conjugative systems in ICEs and plasmids , we should be able to detect a large fraction of ICEs in prokaryotic genomes using information on proteins involved in plasmid conjugation ., Indeed , we checked previously published lists of experimentally studied ICEs 20 , 41 and were able to retrieve all for which experimental validation of mobility by self- conjugation and full sequence data were available ( Table S1 ) ., Two mobilizable elements were missed in our analysis: Tn4555 42 and NBUI1 43 ., These elements are mobilizable and have similar relaxases with no homolog in our genomic bank; as such , we did not include them in our study ., We were thus able to identify all model ICEs in firmicutes ( e . g . Tn916 ) , bacteroides ( e . g . CTnBST ) and proteobacteria ( e . g . SXT , ICEHin1056 , ICEclc ) ., The only exceptions were ICEs of actinobacteria that use FtsK-based transport systems within multi-cellular assemblages ( e . g . pSAM2 ) 44 , 45 ., These systems transport dsDNA not ssDNA between cells within mycelia of some actinobacteria ., As they dont contain relaxases neither T4SS these systems were not expected to be found in our analysis ., Overall , these results indicate that using the accumulated body of knowledge on plasmid conjugation we can extensively identify and class ICEs ., Within the analyzed 1 , 124 complete prokaryotic genomes , which included the 1 , 207 chromosomes and their accompanying 891 PSC plasmids , we identified 335 putative ICEs and 180 putative conjugative plasmids ., Additionally , we found 402 relaxases in chromosomes lacking neighboring T4SS ., If these correspond to IMEs , the estimate of the ratio of conjugative over mobilizable elements both in chromosomes ( ICE/IME\u200a=\u200a0 . 83 ) as in PSC plasmids ( ratio\u200a=\u200a0 . 96 ) is approximately similar and lower than 1 , suggesting that mobilization in trans is frequent in natural populations ., Naturally , mobilization in trans of an IME can only occur if the host genome encodes somewhere else a T4SS with the ability to build a compatible conjugative pilus ., The frequency with which conjugative systems exist in prokaryotic cells is high ., Overall , almost half of the genomes contain a T4SS , either in an ICE ( 18% ) , a conjugative plasmid ( 12% ) or a T4SS without an accompanying relaxase ( 18% ) ., Unfortunately , at this stage we cannot infer computationally if a given T4SS can mobilize another given mobilizable element in trans ., Furthermore , we do not really know how often a T4SS is capable of mobilizing DNA in trans ., Several T4SS that lack neighboring MOB and are involved in protein transport have this ability , e . g . the dot/icm system of Legionella pneumophila 46 ., The Bartonella tribocorum T4SS can also complement deficiencies in the conjugative system of plasmid R388 47 , 48 ., Further experimental work is required to assess the generality of these observations ., An IME or mobilizable plasmid arriving at a cell has a probability of 30% of finding a conjugative element at the time of arrival ., Naturally , given the high flux of these elements , if the mobilizable element remains long enough in the cell it will likely co-reside with a conjugative element ., The probability that a cell harbors a conjugative element at a given moment depends on genome size ( Figure 3 ) ., Small genomes rarely contain ICEs or conjugative plasmids , whereas large genomes often do so ., This fits the common assumption that prokaryotes with smaller genomes engage more rarely in horizontal transfer ., Nevertheless , several small genomes contain conjugative systems , as previously described for Rickettsia 49 and tenericutes 50 ., Some T4SS have been present in the genomes of rickettsiales for a long period of time and their genomic organization is scattered , i . e . conjugation-related genes are not necessarily found in one single cluster 51 ., We used the available literature to annotate these cases 51 ., Analysis of the genomes of other proteobacteria suggests that this situation is relatively rare and that most conjugative systems are coded at one single cluster , which is required to ensure mobility of the locus upon transfer to a new recipient cell ., Using the method explained above we could make the first large-scale quantification of the abundance and diversity of ICEs among prokaryotes ., We found ICEs in all bacterial clades where occurrences have been described previously , including the five major branches ( α , β , γ , ε , δ ) of proteobacteria , the bacteroidetes , and the firmicutes ( Figure 4 ) ., In bacteroidetes , as well as in α- and β-proteobacteria , more than 50% of the available genomes contain at least one ICE ., The other groups show relatively fewer ICEs , with these elements present in less than 30% of the genomes ., We only found one ICE in archaea—in Aciduliprofundum boonei—and one conjugative PSC plasmid—plasmid pNG500 in Haloarcula marismortui ., Yet , we found both in chromosomes and in plasmids many bona fide homologs of VirB4 , often associated with a T4CP ., It is possible that unknown relaxases exist in archaea , since conjugative plasmids are known in this clade and were included in our dataset 52 , 53 ., In actinobacteria , we found many MOB , but few T4SS or T4CP , both in plasmids and chromosomes ., The rarity of T4SS in this clade could be explained by the alternative modes for DNA transfer within mycelia ., Yet , elements in actinobacteria that are classed as mobilizable because they encode a relaxase presumably need a T4CP and T4SS to transfer as we know of no experimental evidence of functional interactions between relaxases and FtsK-based systems ., Therefore , the number of conjugative systems in the clade still seems surprisingly low ., Low sequence similarity is unlikely to be responsible for the lack of identifiable T4SS in actinobacteria since we can uncover distant homologs of VirB4 in all major clades of prokaryotes and we can even indentify by sequence similarity paralogous functionally unrelated ATPases ., We found ICEs and conjugative plasmids in cyanobacteria ., We had previously failed to do so 14 , but the new protein profiles we built are more sensitive and show that this clade also contains conjugative systems both in plasmids and in chromosomes ( to be published elsewhere ) ., Additionally , we found ICEs in acidobacteria , in fusobacteria and one conjugative plasmid in chlorobi ( pPAES01 ) ., In short , all clades with a significant number of sequenced genomes contain conjugative systems showing the ubiquity of this DNA transfer mechanism in the prokaryotic world ., While few ICEs have been experimentally studied in terms of conjugation , we found large numbers of them in the genomes of prokaryotes ., Importantly , we found 86% more ICEs than conjugative plasmids ( Figure 2 and Figure 3 , p<0 . 001 , binomial test ) ., It should be emphasized that this is contrary to the expected if our method was biased , since we use information on plasmid conjugation systems to identify ICEs , not the other way around ., Conjugative plasmids have been most thoroughly studied in proteobacteria whereas ICEs were discovered first in bacteroidetes and in firmicutes 54 ., There is thus often a tendency to consider that conjugative plasmids are prevalent in proteobacteria and ICEs in the other two clades ., Indeed , the preponderance of ICEs over conjugative plasmids varies between clades ( Figure 4 ) ., In firmicutes and bacteroidetes ICEs do represent respectively 84% and 81% of all conjugative elements , while in proteobacteria ICEs only slightly outnumber conjugative plasmids ., We identified no conjugative PSC plasmid within actinobacteria ., Cyanobacteria were the only clade for which we found more conjugative plasmids ( 11 ) than ICEs ( 4 ) ., While we found conjugative plasmids in several different genera of cyanobacteria ( Cyanothece , Nostoc , Anabaena , Acaryochloris ) , we only found ICEs in Cyanothece ., Besides confirming the preconception that , in bacteroidetes and firmicutes , ICEs outnumber conjugative plasmids , we show that prevalence of ICEs over conjugative plasmids is almost general ., ICEs might be more abundant in the analyzed genomes because of sequencing biases ., First , certain sequencing projects might have ignored the sequencing of plasmids ., Second , if ICEs are more stable in genomes than plasmids , bacterial culturing might induce a bias towards the over-representation of ICEs ., In any case , our results clearly demonstrate that ICEs are a significant fraction of all conjugative elements in prokaryotes ., We next investigated if conjugation systems in plasmids and ICEs are of similar types ., For this , we divided the conjugative systems found in proteobacteria into the four different archetypes: MPFF , MPFT , MPFI and MPFG ., MPFT conjugative pili are short and thick , mate essentially in solid media and include elements such as CTn4371 55 and MlSymR7A 56 ., MPFT are equally distributed , in relative terms , among conjugative plasmids and ICEs ( Figure 4 ) ., Interestingly this is not the case for the other mating types that show significantly different frequencies among plasmids and ICEs ( p<0 . 001 , χ2 test ) ., MPFF , which have long flexible pili , mate efficiently in solid and liquid , and include the SXT family 39 ., These pili are rare among ICEs , whereas they are the second most frequent type in plasmids ., On the other hand , the MPFG pili have only been described to mate in solid surfaces 18 and are found essentially among ICEs , e . g . the clc or pKLC102 elements of Pseudomonas 57 , 58 ., We found few MPFI systems in plasmids and even fewer in chromosomes ., The latter were essentially found in the dot/icm systems of Legionella and Coxiella , where only the latter encode a MOB close to the T4SS ., As a result , MPF types known to mate in liquid are under-represented in ICEs relative to plasmids ., We then analyzed the co-occurrence of ICEs in a given genome ., Conjugative plasmids rarely code for two T4SS and , when they do , they tend to have multiple MPFT 14 ., We found 73 chromosomes encoding multiple ICEs and 32 genomes containing multiple conjugative plasmids ., We found all MPF types in multiple copies , except for MPFI in chromosomes and MPFG in plasmids , but this could result from their rarity ., A striking previously described case concerns Orientia tsutsugamushi genomes , which contain a large number of conjugation-related genes in clusters that for the most part present evidence of pseudogenization 59 ., It is unclear in this case how many effective conjugation systems are encoded in the chromosome , but we could identify 5 complete clusters of MPFF ., In our dataset the largest number of intact ICEs ( seven ) was found in Bordetella petrii DSM 12804 ( which comprises both MPFT and MPFG elements ) and in the firmicute Clostridium difficile 630 ., The genome of Agrobacterium vitis S4 contains the largest number of conjugative plasmids ( 4 , all MPFT ) ., In summary , conjugative systems in chromosomes and plasmids co-occur and sometimes in large numbers ., This is expected , since each ICE is an independent element ., This suggests that different types of T4SS can co-exist in a functional state in the cell ., Discrimination between T4SS could be achieved by the specificity of the T4CP ., Alternatively , one could imagine that in some cases conjugative elements also use T4SS encoded in trans ., One major surprising finding of this work was the high number of T4SS lacking nearby relaxases and thus not classed as ICEs ( Figure 2 ) ., We can explain these findings in three different ways: as an artifact , as an indication of unknown relaxases or as evidence of high frequency of T4SS not involved in conjugation ., Artifacts can occur in our analysis in several ways ., First , one might have found many false positives in the detection of VirB4 ., This is unlikely because in proteobacteria ( 33% of MOBless T4SS ) , we find MOBless virB4 genes neighboring other type-specific genes of T4SS ( 92 out of 109 clusters ) ., This shows that in the vast majority of cases the virB4 assignment in MOBless T4SS is correct ., In the 17 remaining cases we almost always find at least one T4SS specific gene neighboring the MOBless virB4 gene ( 16 out of 17 cases ) , but not enough to make it a valid cluster , suggesting that these loci correspond to inactive T4SS ongoing genetic degradation ., Second , we might be failing to identify a large number of homologous T4CP or MOB in conjugative systems and this might lead to the misclassification of these clusters as MOBless T4SS ., Yet , this does not fit the remaining observations: that MOBless T4SS are much more abundant in chromosomes than in plasmids and that we are able to identify VirB4 , T4CP and MOB in clades distant from proteobacteria ., All these pieces of evidence advocate against the hypothesis that the large number of MOBless T4SS is a consequence of methodological artifacts ., We showed above that the abundance of ICEs and conjugative plasmids depends strongly on genome size and that small genomes are practically devoid of conjugative systems ( Figure 3 ) ., The distribution of MOBless T4SS is very different since these elements are abundant in small genomes and their frequency practically does not change with genome size ( Figure 3 ) ., Small genomes tend to correspond to bacterial pathogens , and many of these are known to use T4SS to secrete proteins into the host cells for their subversion ., T4SSs used for protein transport , as opposed to conjugation , have been described in strains of Bartonella , Brucella , Bordetella , the Legionellales , Helicobacter , and the Rickettsiales 46 , 60–64 ., Out of the 109 MOBless T4SS in proteobacteria , 77 are indeed found among these clades reinforcing the speculation that MOBless T4SS do often correspond to protein secretion systems ., If so , this would include MPFF elements , not known before to be recruited for that , and several clades of environmental prokaryotes , which so far were not known to carry such protein transport systems ., We have not yet done the precise delimitation of ICEs in genomes ., Yet , we already carried out a preliminary analysis of the integrases co-localizing with the T4SSs to check for differences between ICEs and MOBless T4SSs ., As described above , most ICEs include a tyrosine or serine recombinase and only a minority of well-characterized elements integrate by other means ., Therefore the conjugation systems we identify in genomes are expected to have neighboring integrases ., Co-localization of MOBless T4SS with integrases is expected under a number of situations:, ( i ) if the protein secretion system is in a mobile element itself , as is frequently the case for T3SS 65 , 66;, ( ii ) if it represents an element undergoing genetic degradation is which the relaxase was inactivated but not the integrase nor the T4SS genes;, ( iii ) or if the genes encoding the T4SS happen to be near an unrelated mobile element ., Yet , since integration is strictly necessary for ICE , we did expect to find more integrases neighboring the T4SS of ICE than those of MOBless T4SS ., Using the PFAM domains ( PF00589 for the tyrosine recombinases; PF07508 and PF00239 for serine recombinases ) , we found that within proteobacteria 87% of the ICEs and 50% of the MOBless T4SSs have a neighboring integrase distant no more than 60 genes from the conjugation-related genes ., The difference is highly significant ( p<0 . 001 , binomial test ) and suggests that MOBless T4SS are indeed intrinsically different from ICEs ., We then analyzed the other clades to see if their MOBless T4SS were more frequently neighboring integrases since that could be the sign of the presence of unnoticed relaxases in these poorly studied genomes ., We found that 90% of the ICEs and 56% of the MOBless T4SS in these other clades contain an integrase , within a distance of less than 60 genes , which is very close to the values found in proteobacteria ., These results are consistent with intrinsic functional differences between the T4SS of ICEs and the MOBless T4SS ., Finally , we analyzed the co-occurrence of relaxases with T4SS in ICEs ( Figure 5 ) ., Many MOB/MPF combinations are found among conjugative elements ., This suggests that the MOB and MPF modules can shuffle over long evolutionary distances ., However , there are some expected relevant associations between MPF and MOB , e . g . MPFT with MOBP or MPFF with MOBF as suggested by their frequent association in conjugative plasmids 14 , 67 ., Among less studied groups , MOBB is specific of bacteroidetes and MPFG only use one type of relaxase , MOBH ( 58 cases in chromosomes and 2 in plasmids ) ., It is therefore possible that some sub-types of T4SS use yet unknown relaxases ., In particular , it is tempting to suggest that this is the case in archaea where we find very few relaxases ., As conjugation is an agent of horizontal transfer , and some very broad range plasmids have been described , one might expect little concordance between the phylogeny of VirB4 and that of the 16S rDNA ., Yet , in plasmids it was found that large clades within bacteria corresponded to large clades in VirB4 with little apparent transfer between domains 14 ., To check that similar results are still valid when using the information on ICEs and the new data on cyanobacteria and bacteroidetes , we made a phylogenetic analysis of the only ubiquitous element of T4SS: VirB4 ( see Materials and Methods ) ., This tree was built using a non-redundant subset of proteins and shows several remarkable things ( Figure 6 ) ., First , MPF classification within proteobacteria remains meaningful , since the four types ( F , G , I , T ) are found in four monophyletic groups that exhibit strong support values ., Both cyanobacteria and bacteroidetes form monophyletic clades , suggesting lack of significant transfer of conjugative systems between these and other clades since their divergence ., This is consistent with their specific relaxases: MOBV is mainly found in cyanobacteria and MOBB is only found in bacteroidetes ( Figure 5 ) ., Firmicutes and actinobacteria ( FA in Figure 6 ) , on one side , and firmicutes , actinobacteria , tenericutes and archaea ( FATA in Figure 6 ) , on the other , form the two remaining clades , but inside these groups one still finds mostly monophyletic clades ., Thus , while elements propagating by means of conjugation systems are the most promiscuous known agents of horizontal transfer , the evolution of these systems does not show signs of frequent transfer of mobility backbone modules between types ., The existence of every type of T4SS in both chromosomes and plasmids of proteobacteria , albeit at very diverse frequencies , suggest that conjugative plasmids and ICEs have exchanged T4SS along their evolutionary history ., To test this , we marked in the phylogenetic tree of VirB4 the respective genes that were encoded in chromosomes and in plasmids ., An example for the MPFT is presented in Figure 6 ., If ICEs were derived from conjugative plasmids , then one would expect large monophyletic clades of ICEs , indicating creation of the ICE , and clades devoid of ICEs , indicating lack of creation within the lineage ., Furthermore , one would see evidence of plasmids as ancestral traits in the tree ., If conjugative plasmids were derived from ICEs then the opposite picture should arise ., The data presented in this work is not suggestive of any of these scenarii ., Conjugative plasmids and ICEs ( or chromosomal T4SS lacking nearby MOB ) are intermingled along the whole tree ( data not shown ) ., At closer phylogenetic distances , i . e . the comparisons including the 15% of the tree closest to the tips , we do observe that the most similar VirB4 of an ICE is in general a VirB4 from another ICE and the reciprocal occurs for conjugative plasmids ( Figure 7 ) ., We found 5 pairs of VirB4 encoded in different types of replicons that are distant by less than 1% in the tree ., In three of the cases they are in a chromosome of one species and in a plasmid of another species within enterobacteria ., Hence , at short evolutionary distances , plasmids and ICEs are indeed distinguishable ., Yet , at slightly larger distances this signal quickly disappears and the ICEs and conjugative plasmids are perfectly mixed ., The resulting picture is that one finds ICEs resembling much more some conjugative plasmids than other ICEs ., For the most part of the evolutionary history of conjugation , ICEs have probably been converted to and from plasmids ., As conjugative systems of both plasmids and ICEs shared most of their evolutionary history , they should be regarded as one and the same ., In this work we present the results of a semi-automatic method to detect conjugation-associated mobility systems not only in plasmids but also in chromosomes ., This analysis paves the way for a systematic quantification of conjugation systems in prokaryotic genomes and in metagenomic data ., When coupled with the detection of integration junctions ( work in progress ) it will also allow to analyze the gene repertoires of ICEs , and evaluate the evolutionary interplay between ICEs , conjugative plasmids and phages ., Therefore , our present results only concern the C part of ICEs and conjugative plasmids ., In the case of ICEs , this only gives an indication of their position in genomes , but not of their limits ., ICEs can be very large ( more than 500 kb for ICEMlSymR71 of Mesorhizobium loti 56 ) ., Since the size of the C part is more or less constant , the variations in ICEs size will reveal the cargo genes they contain , much like for plasmids ., The next step of this work will thus be to delimit ICEs within genomes in order to study the genes they carry ., Our quantitative analysis shows that conjugative systems are more likely to be found in larger genomes ., This fits the current assumption that larger genomes engage more frequently in horizontal gene transfer ., The study of the cargo genes will help to quantify and qualify the role of ICEs in the functional diversification of prokaryotes ., Our analysis of MOBless T4SS in proteobacteria strongly suggests that many of t

Introduction, Results, Discussion, Materials and Methods

Horizontal gene transfer shapes the genomes of prokaryotes by allowing rapid acquisition of novel adaptive functions ., Conjugation allows the broadest range and the highest gene transfer input per transfer event ., While conjugative plasmids have been studied for decades , the number and diversity of integrative conjugative elements ( ICE ) in prokaryotes remained unknown ., We defined a large set of protein profiles of the conjugation machinery to scan over 1 , 000 genomes of prokaryotes ., We found 682 putative conjugative systems among all major phylogenetic clades and showed that ICEs are the most abundant conjugative elements in prokaryotes ., Nearly half of the genomes contain a type IV secretion system ( T4SS ) , with larger genomes encoding more conjugative systems ., Surprisingly , almost half of the chromosomal T4SS lack co-localized relaxases and , consequently , might be devoted to protein transport instead of conjugation ., This class of elements is preponderant among small genomes , is less commonly associated with integrases , and is rarer in plasmids ., ICEs and conjugative plasmids in proteobacteria have different preferences for each type of T4SS , but all types exist in both chromosomes and plasmids ., Mobilizable elements outnumber self-conjugative elements in both ICEs and plasmids , which suggests an extensive use of T4SS in trans ., Our evolutionary analysis indicates that switch of plasmids to and from ICEs were frequent and that extant elements began to differentiate only relatively recently ., According to the present results , ICEs are the most abundant conjugative elements in practically all prokaryotic clades and might be far more frequently domesticated into non-conjugative protein transport systems than previously thought ., While conjugative plasmids and ICEs have different means of genomic stabilization , their mechanisms of mobility by conjugation show strikingly conserved patterns , arguing for a unitary view of conjugation in shaping the genomes of prokaryotes by horizontal gene transfer .

Some mobile genetic elements spread genetic information horizontally between prokaryotes by conjugation , a mechanism by which DNA is transferred directly from one cell to the other ., Among the processes allowing genetic transfer between cells , conjugation is the one allowing the simultaneous transfer of larger amounts of DNA and between the least related cells ., As such , conjugative systems are key players in horizontal transfer , including the transfer of antibiotic resistance to and between many human pathogens ., Conjugative systems are encoded both in plasmids and in chromosomes ., The latter are called Integrative Conjugative Elements ( ICE ) ; and their number , identity , and mechanism of conjugation were poorly known ., We have developed an approach to identify and characterize these elements and found more ICEs than conjugative plasmids in genomes ., While both ICEs and plasmids use similar conjugative systems , there are remarkable preferences for some systems in some elements ., Our evolutionary analysis shows that plasmid conjugative systems have often given rise to ICEs and vice versa ., Therefore , ICEs and conjugative plasmids should be regarded as one and the same , the differences in their means of existence in cells probably the result of different requirements for stabilization and/or transmissibility of the genetic information they contain .

bacteriology, genomics, genome evolution, heredity, genetics, molecular genetics, biology, gene flow, microbiology, bacterial evolution, genetics and genomics

journal.pgen.1005182

Ataxin-2 Regulates RGS8 Translation in a New BAC-SCA2 Transgenic Mouse Model

Spinocerebellar ataxia type 2 ( SCA2 ) belongs to the group of neurodegenerative diseases caused by polyglutamine ( polyQ ) expansion ., This group includes SCA1 , Machado-Joseph disease ( SCA3 or MJD ) , SCA6 , SCA7 , SCA17 , Huntingtons disease , spinal bulbar muscular atrophy ( SBMA ) and dentatorubral-pallidoluysian atrophy ( DRPLA ) ., SCA2 is an autosomal dominant disorder leading to motor incoordination which is caused by progressive degeneration of cerebellar Purkinje cells , and selective loss of neurons within the brainstem and spinal cord 1 ., As with most autosomal dominant ataxias , symptoms are characterized by a progressive loss of motor coordination , neuropathies , slurred speech , cognitive impairment and loss of other functional abilities arising from Purkinje cells and deep cerebellar nuclei 2 , 3 ., In SCA2 , expansion of a CAG repeat in exon 1 of the Ataxin-2 ( ATXN2 ) gene causes expansion of a polyQ domain in the ATXN2 protein ., As in the other polyQ diseases , the length of the polyQ repeat is inversely correlated with age of onset ( AO ) in SCA2 1 , 4 ., In contrast to other polyQ diseases , mutant ATXN2 does not enter the nucleus in appreciable amounts in early stages of disease ., This is also confirmed by protein interaction studies that have identified ATXN2 interactors with cytoplasmic localization 5–8 ., Polyglutamine disorders show their pathology through a toxic gain of function of the protein and larger polyQ expansions have been associated with greater pathology 3 , 9 ., ATXN2 is widely expressed in the mammalian nervous system 1 , 10 , 11 ., It is involved in regulation of the EGF receptor 12 , and the inositol 1 , 4 , 5-triphosphate receptor ( IP3R ) whereby increased cytosolic Ca2+ occurs with CAG repeat expansion 13 ., ATXN2 functions are also associated with the endoplasmic reticulum 14 , and the Golgi complex 15 ., Studies in Caenorhabditis elegans support a role for ATXN2 in translational regulation as well as embryonic development 6 ., ATXN2 is also important in energy metabolism and weight regulation , as mice lacking Atxn2 , developed obesity and insulin resistance 16 , 17 ., Furthermore , ATXN2 interacts with multiple RNA binding proteins , including polyA binding protein 1 ( PABP1 ) , the RNA splicing factor A2BP1/Fox1 , DDX6 , TDP-43 , and has been localized in polyribosomes and stress granules demonstrating its unique role in RNA metabolism 5 , 6 , 8 , 18 ., Several SCA2 mouse models have been generated ., We have reported two transgenic mouse models in which expression of full-length ATXN2 with 58 or 127 CAG repeats ( ATXN2-Q58 or ATXN2-Q127 ) is targeted to Purkinje cells ( PCs ) using the Purkinje cell protein-2 ( Pcp2 ) promoter 19 , 20 ., These lines show progressive motor phenotypes accompanied by the formation of insoluble cytoplasmic aggregates , loss of PCs , and shrinkage of the molecular layer associated with the reduction of calbindin staining in PC bodies and dendrites ., Onset of the motor phenotype of Pcp2-ATXN2Q127 mice is associated with reduced PC firing that is progressive with age 20 ., Another Atxn2-CAG42 knock-in mouse model demonstrated very late-onset motor incoordination associated , but this was seen only in homozygous knock-in animals ., This was associated with Pabpc1 deficiency , and upregulation of Fbxw8 , but without loss of calbindin staining or downregulation of Calb1 mRNA 21 ., In order to model human diseases using cis-regulatory elements , recent mouse and rat models have been created by transgenesis using human bacterial artificial chromosomes ( BACs ) 22–27 ., In the BAC approach , an entire human gene including introns and regulatory regions is introduced into the mouse genome ., BAC models often have lower genomic copy numbers than conventional cDNA transgenic models resulting in more physiological expression levels and a potentially more faithful late onset of disease ., We developed new BAC-SCA2 transgenic mouse lines expressing full-length human wild-type or mutant ATXN2 genes including upstream and downstream regulatory sequences ., BAC mice with mutant ATXN2 exhibited progressive neurological symptoms and morphological changes in cerebellum ., We used this mouse model to confirm changes in key PC-genes identified in a cDNA transgenic model , in particular the effects of mutant ATXN2 on Rgs8 steady state protein levels ., To understand the pathological and behavioral effects in the context of physiologic expression of human wild-type and mutant ATXN2 , we engineered a 169 kb human BAC ( RP11-798L5 ) that contained the entire 150 kb human ATXN2 locus with 16 kb of the 5’ flanking genomic sequence and 3 kb of the 3’ flanking genomic sequence ( Fig 1A ) ., The authenticities of these constructs were subsequently verified by Southern blot and restriction site analyses ( S1 Fig ) ., The CAG tract was mutation-free when sequenced from both strands ., After transgenic microinjection of purified intact BAC DNAs , one line each for control ( BAC-ATXN2-Q22 ) and one for mutant mice ( BAC-ATXN2-Q72 ) was further analyzed ., These lines will be designated as BAC-Q22 and BAC-Q72 in the remainder of the text ., Quantitative PCR ( qPCR ) analyses of genomic DNA revealed that both BAC-Q22 and BAC-Q72 mice had tandem integrates of 10 and 4 copies of the ATXN2 transgene , respectively ., In RT-PCR analyses , both BAC-Q22 and BAC-Q72 mice demonstrated the expression of intact human ATXN2 transcripts throughout the central nervous system ( CNS ) , including cerebral hemispheres , cerebellum and spinal cord ( Fig 1B ) ., Non-CNS tissues , including heart and liver also showed ATXN2 transgene expression ( Fig 1B ) ., The authenticities of PCR products were confirmed by sequencing ., We further determined relative expression of ATXN2 transcripts in the two BAC transgenic lines by quantitative RT-PCR ., BAC-Q22 cerebella had higher expression of human ATXN2 than BAC-Q72 cerebella while the expression of endogenous mouse Atxn2 remained unchanged in both compared with wild-type mice ( Fig 1C ) ., To assess protein expression , we performed Western blot analysis using cerebellar extracts of 16 week-old animals and a monoclonal antibody ( mAb ) to human ATXN2 ., The results showed that BAC mice expressed full-length human wild-type or mutant ATXN2 protein ., Of note , protein levels of ATXN2-Q22 were higher than those of ATXN2-Q72 ., Furthermore , we confirmed the ATXN2-Q72 protein expression using 1C2 mAb , an antibody against an expanded polyQ epitope in Western blot analyses ( Fig 1D ) ., These results demonstrate that human ATXN2 transgenes ( ATXN2-Q22 and ATXN2-Q72 ) were properly expressed in BAC mice ., In addition to ATXN2 , three overlapping genes ( U7 . 1–202 snRNA , RP11-686G8 . 1–001 and RP11-686G8 . 2–001 ) are contained in the human BAC ., Quantitative RT-PCR analyses of wild-type and BAC transgenic mouse cerebellar RNAs demonstrated that the relative expression of each overlapping gene to that of the ATXN2 transgene did not differ between BAC-Q22 and BAC-Q72 animals indicating these genes did not contribute to the phenotypes associated with CAG expansion in the ATXN2 gene ( S2 Fig ) ., The Allen Brain Atlas shows widespread expression of human ATXN2 with very significant expression levels in the cerebellum 28 ., Given the nature of ATXN2 expression in brain , we determined the expression of human ATXN2 transgene transcript in sub-regions of mouse brain including spinal cord using qRT-PCR ., Expression of endogenous mAtxn2 was evident in many regions including frontal , occipital and olfactory cortex , hippocampus , thalamus , basal ganglia , cerebellum and spinal cord ., Human ATXN2 transgene expression was found in all regions tested , but relatively higher expression was observed in the basal ganglia ( S3 Fig ) ., As cerebellar degeneration is predominant in SCA2 , we further examined the expression patterns of the ATXN2 transgene in discrete areas of the cerebellum using laser-capture microdissection ( LCM ) ., We captured molecular layer ( ML ) , Purkinje cells ( PCs ) , granule cell layer ( GCL ) and dentate nuclear ( DN ) fractions ., Relative enrichment was determined by measuring expression of a cell-type specific marker genes using qRT-PCR ., Evidence for expression of endogenous mAtxn2 was found in all fractions , but was highest in Purkinje cells ., Expression of transgenic ATXN2 was also seen in all fractions , although small differences in expression levels existed between BAC-Q22 and BAC-Q72 ( Fig 2A and 2B ) ., LCM was remarkably successful in separating cerebellar neuronal population as shown by expression of marker genes for PCs and molecular layer ( Pcp2 and Calb1 ) , granule cells ( Neurod1 ) and dentate neurons ( Spp1 ) ( Fig 2C and 2F ) ., In summary , inclusion of regulatory regions in the human BAC transgene led to expression of the transgene that mirrored expression of mouse Atxn2 including low but detectable expression in GCs and DNs ., By visual inspection both BAC transgenic lines ( BAC-Q22 and BAC-Q72 ) had a smaller body size than wild-type littermates beginning at 8 weeks of age ., By 24 weeks of age , both BAC transgenic mice weighed about 30% less than their wild-type littermates ( Wild-type = 33 . 9 ±3 . 8; BAC-Q22 = 24 . 6 ±3 . 6 and Wild-type = 32 . 1 ±2 . 8; BAC-Q72 = 22 . 9 ±3 . 7 ) ., BAC-Q72 mice did not show an abnormal home cage behavior ., To assess the development of motor impairment , both BAC transgenic lines and wild-type littermates were tested using the accelerating rotarod paradigm at several time points ( Fig 3 ) ., BAC-Q22 mice performed as well as wild-type littermates at 8 , 16 and 36 weeks of age ( Fig 3 ) suggesting that expression of wild-type human ATXN2 was not detrimental to motor function ., BAC-Q72 mice were tested at 5 , 16 and 36 weeks of age and compared with their wild-type littermates ., BAC-Q72 mice showed normal performance at 5 weeks ( Fig 3 ) and at 12 weeks ( S4A Fig ) ., Of note , testing at 12 weeks was performed on mice housed under slightly different conditions , which may explain the relatively poor performance of wild-type mice ., At 16 weeks of age , performance of BAC-Q72 mice became significantly worse than wild-type mice ( Fig 3; p<0 . 05 ) and mice continued to perform poorly as they aged ( 24 and 36 weeks old , S4A Fig and Fig 3 ) ., Taken together , these results indicate that BAC-Q72 transgenic mice develop a progressive age-dependent motor impairment ., To investigate morphological changes associated with the expression of mutant ATXN2 protein , we compared cerebellar sections from BAC transgenic lines with wild-type mice ., Immunostaining with calbindin-28k antibody revealed PC morphological changes in BAC-Q72 mice at 24 weeks of age , but not in BAC-Q22 or wild-type mice ( Fig 4A ) ., To more quantitatively assess this change , we performed Western blotting and verified reduction of Calb1 and Pcp2 proteins in BAC-Q72 mouse cerebella ( Fig 4B ) ., As observed in the Pcp2-ATXN2Q127 model , cerebellar morphology was still normal at a time when key mRNA transcripts had already declined ., Thus , calbindin-stained cerebellar sections and PC counts of BAC-Q72 mice at 12 weeks showed normal cerebellar morphology and unaltered PC counts 18 . 8 ±1 . 2 in WT , n = 3 animals , and 19 . 4 ±1 . 1 in BAC-Q72 mice , n = 3 animals , p = 0 . 51 ( S4B , S4C Fig ) ., We previously reported that steady-state mRNA levels of specific PC transcripts preceded behavioral onset in an SCA2 model targeting transgene expression to PCs 20 ., Expression changes in these genes ( Calb1 , Pcp2 , Grid2 and Grm1 ) also preceded the onset of a decrease in PC firing ., Expression changes were progressive over time and paralleled deterioration of motor behavior ., To investigate whether similar changes occurred in BAC transgenic mice as we previously observed in Pcp2-ATXN2Q127 , we performed qRT-PCR to measure transcript levels of PC-specific genes at different ages ., At 16 and 45 weeks , BAC-Q22 mice were indistinguishable from wild-type mice including expression of endogenous mouse Atxn2 ( Fig 5A ) ., In BAC-Q72 mice , however , expression of Pcp2 showed significant reductions ( p<0 . 01 ) as early as 5 weeks ., All other genes tested remained unchanged compared to wild-type ( Fig 5B ) ., At 9 and 16 weeks of age , significant reductions in Calb1 ( p<0 . 05 ) and Grid2 ( p<0 . 01 ) were seen and were progressive ( Fig 5B ) ., Steady-state levels of Grm1 decreased only at 24 weeks ( p<0 . 05 ) ., Endogenous mouse Atxn2 expression levels did not change in BAC-Q72 mice at any time point when compared with wild-type ., Taken together , these data demonstrated that a subset of PC-enriched genes showed a progressive reduction in steady-state mRNA levels in BAC-Q72 mice , whereas they remained unchanged in BAC-Q22 animals ., To further characterize the BAC-Q72 line and compare it with the well-characterized Pcp2-ATXN2Q127 line , we performed transcriptome analysis by deep RNA-sequencing of cerebellar RNA ., We chose time points for both lines just prior to behavioral and morphological changes , i . e . 8 weeks for the BAC-Q72 line and 6 weeks for the Pcp2-ATXN2Q127 line ., For both sets of RNAs , quality of reads and alignments were high ( see methods ) ., We observed significant changes of 1417 transcripts in Pcp2-ATXN2Q127 and 491 transcripts in BAC-Q72 mice with a false discovery rate ( FDR ) of ≥15 and a log2 ratio of change ≥|0 . 30| ( Fig 6A ) ., With these filtering parameters , 255 transcripts were only seen in the BAC-Q72 line ( class I ) , 236 transcripts were shared between the two lines ( class II ) and 1181 transcripts were changed only in the Pcp2-ATXN2Q127 line ( Class III ) ., We validated changes in several of the class II transcripts by qRT-PCR using cerebellar RNA samples from BAC-Q72 mice ( 8 weeks old ) and Pcp2-ATXN2Q127 ( 6 weeks old ) , and compared with their respective WT littermates ( Fig 6B ) ., The concordance between RNA-seq and qRT-PCR was high ( Fig 6C ) ., The top 50 transcripts changed in the BAC-Q72 line are shown in S1 Table and the top 50 transcripts changed in the Pcp2-ATXN2Q127 line are presented in S2 Table ., This table also shows that most of these transcripts are changed in the BAC-Q72 line as well , although with a smaller degree of change or a lower FDR ., S3 Table lists the top class II genes sorted by FDR in the BAC-Q72 line ., This represents a subset of the 236 overlapping genes shown in Fig 6A ., In order to gain insight into the molecular function of altered transcripts in BAC-Q72 and Pcp2-ATXN2Q127 mice , we performed Gene Ontology ( GO ) analysis ., This is shown in S4 Table and indicates that many of the significant GO terms are shared by the two models ., Of note , GO terms relate to known functions of PC such as calcium homeostasis , glutamate-mediated signaling and voltage-gated ion channels ., In summary , these data indicate a significant overlap of altered transcripts and shared functions in both SCA2 models at comparable stages just prior to onset of morphological and behavioral changes ., We were also interested in the nature and expression pattern of transcripts in class I and class III ( Fig 6 ) ., We confirmed changes in several of the class I transcripts by qRT-PCR ( S5 Fig ) ., These transcripts showed a progressive reduction in BAC-Q72 mice , but remained unchanged in the Pcp2-ATXN2Q127 line even at late time points ., Of these 50 , 16 genes ( Grm4 , Igfbp5 , Fstl5 , Snrk , D8Ertd82e , Dusp5 , Nab2 , Btg1 , Adrbk2 , Slc25a29 , Sty12 , Crhr1 , Synpr , Lrrtm2 , Rit2 and Cabp2 ) were previously identified as GC-specific using translational profiling 29 ., Class III transcripts were those that showed changes only in Pcp2-ATXN2Q127 mice , but not in BAC-Q72 at an FDR>15 and a log2 ratio of change ≥|0 . 3| ., We verified expression changes of six class III transcripts longitudinally in Pcp2-ATXN2Q127 mice at 4 , 8 , and 24 weeks of age , and BAC-Q72 mice at 5 , 9 , 16 and 24 weeks of age , and their respective WT littermates by qRT-PCR ., Five of the six transcripts showed significant and progressive reduction with age not only in Pcp2-ATXN2Q127 mice but also in BAC-Q72 mice ( S6 Fig ) ., This is consistent with the milder behavioral phenotype seen in BAC-Q72 mice and suggests that the overlap of the transcriptomes in the two models may potentially be even greater ., Changes in steady-state expression of a subset of genes preceded onset of physiological and behavioral changes in Pcp2-ATXN2Q127 and BAC-Q72 mice ., One of the most significantly down-regulated genes in both models prior to behavioral onset was Rgs8 ( regulator of G-protein signaling 8 ) ( S1 , S2 , S3 Tables ) ., RGS proteins are regulatory and structural components of G protein-coupled receptor complexes ., RGS proteins ( RGS7 , RGS8 , RGS11 , RGS17 and RGSz1 ) are widely expressed in cerebellum and RGS8 is specifically distributed in dendrites and cell bodies of PCs 30 , 31 ., Several reports suggest that the RGS family proteins are also associated with motor neuron functions 32 , 33 ., The decreased steady-state level of Rgs8 mRNA was confirmed by qRT-PCR in Pcp2-ATXN2Q127 mice at 4 , 8 and 24 weeks of age , indicating that these RNAs progressively declined with time ( S7A Fig ) ., In parallel , we also measured Rgs8 protein steady state levels in Pcp2-ATXN2Q127 mouse cerebella at 24 weeks of age ., As expected , Rgs8 protein levels were significantly reduced in Pcp2-ATXN2Q127 mice when compared with wild-type mice ( S7B Fig ) ., Next , we investigated the fate of Rgs8 mRNA steady-state levels in our BAC mouse models by qRT-PCR ., When tested in BAC-Q72 mouse cerebella , levels of Rgs8 mRNA progressively decreased with time but remained unchanged in BAC-Q22 mice compared with wild-type mice across all ages of mice tested ( Fig 7A ) ., To examine whether changes in steady-state mRNA levels led to decreased protein abundance , we performed Western blot analysis to measure Rgs8 protein in wild-type and BAC transgenic mouse cerebella ., Western blot analyses indicated reduced steady-state levels of Rgs8 protein in BAC-Q72 mice but not in BAC-Q22 mice when compared with wild-type mice at 24 weeks of age ( Fig 7B ) ., To assess whether these findings replicated in human cells we analyzed EBV-transformed lymphoblastoid ( LB ) cells derived from a control individual and two SCA2 patients with expansions of Q46 and Q52 ( Fig 7C ) ., We could not use skin fibroblasts as this cell type does not express RGS8 ., Two SCA2-LB cells expressing Q46 or Q52 demonstrated decreased expression of RGS8 transcript compared with control cells expressing wild-type ATXN2 with 22 repeats ., Unfortunately , LB cells do not efficiently translate RGS8 message , so that Western blots did not allow detection of RGS8 protein in LB cells ., To test whether reduction of Rgs8 levels induced by mutant ATXN2 could be recapitulated in vitro , we measured steady-state levels of RGS8 mRNA and protein in hygromycin selected enriched SH-SY5Y cells expressing Flag-tagged ATXN2-Q22 , -Q58 or -Q108 ., Western blot analyses of whole cell extracts indicated that expression of ATXN2-Q58 or Q108 resulted in decreased RGS8 levels compared to control or ATXN2-Q22 ( Fig 8A ) ., To exclude that decreased RGS8 levels were a consequence of selective cellular toxicity of ATXN2-Q58 or -Q108 expression , we measured expression of endogenous DDX6 and PABPC1 , which have been shown to interact with ATXN2 6 , 8 and CUG-BP1 , a nuclear protein by Western blot analysis ., The levels of DDX6 , PABPC1 and CUG-BP1 were not altered ( Fig 8A ) strongly supporting that the effect of mutant ATXN2 was specific to RGS8 ., In parallel , qRT-PCR analyses of SH-SY5Y cell lines expressing Flag-tagged wild-type and mutant ATXN2 demonstrated a moderate reduction of RGS8 mRNA in cell expressing Flag-ATXN2-Q108 ( Fig 8B ) ., Decrease of RGS8 levels in mutant BAC mice could be the result of transcriptional control , mRNA stability and processing or translational control ., In contrast to other polyQ proteins , ATXN2 does not enter the nucleus 19 and protein interaction studies have not yielded proteins thought to be involved in transcriptional control ., To examine translation of RGS8 , we expressed exogenous RGS8 in hygromycin selected SH-SY5Y cells expressing Flag-tagged ATXN2-Q22 , -Q58 or -Q108 ., MYC-tagged RGS8 cDNA including 5’ and 3’ UTRs was cloned under the transcriptional control of the CMV promoter ., Forty-eight hrs post-transfection , Western blot analyses revealed that the levels of exogenous RGS8 were significantly decreased in cells expressing ATXN2-Q58 or -Q108 compared with cells expressing wild-type ATXN2-Q22 ( Fig 8C ) ., To control for equal transfection , we monitored levels of GFP , which was expressed as an independent cassette in the plasmid ., Thus , presence of mutant ATXN2 reduced RGS8 protein levels in vivo and in vitro ., Reduced protein levels potentially out of proportion to reduced mRNA levels in vivo and in vitro suggested to us that ATXN2 might be directly involved in the translation or stability of specific mRNAs ., In addition , ATXN2 is known to interact with RNAs through a “Like Sm ( LSm ) domain” 34–36 ., It also interacts with cytoplasmic poly ( A ) -binding protein 1 ( PABPC1 ) and assembles with polysomes 6 , 7 ., Therefore , we first tested interaction of ATXN2 with RGS8 mRNA and then performed in vitro translation assays in the presence of wild-type and mutant ATXN2 ., We performed Protein-RNA immunoprecipitation ( IP ) experiments in cultured SH-SY5Y cells overexpressing Flag-tagged ATXN2 containing Q22 or Q108 ., Whole cell extracts were incubated with Flag-mAb-beads and immunoprecipitates were washed with a buffer containing 200 mM NaCl ., Bound protein-RNA complexes were eluted from the beads by Flag peptide competition ., The IP products were divided equally into two aliquots and one aliquot was analyzed by Western blot ., As shown in Fig 9A , the eluted proteins showed co-IP of DDX6 and PABPC1 , which are known to interact with ATXN2 6 , 8 ., To identify RNAs that immunoprecipitated with ATXN2 , the extracted RNAs from the second aliquot were subjected to RT-PCR and qPCR analyses ., Our results showed that RGS8 mRNA precipitated with ATXN2-Q22 and ATXN2-Q108 ( Fig 9A and 9B ) ., Binding of RGS8 mRNA with ATXN2-Q108 , however , was significantly reduced compared with ATXN2-Q22 in three independent experiments ., We next proceeded to examine in vitro RGS8 translation ., For that purpose , we performed assays using Flag-tagged ATXN2 with Q22 or Q108 , respectively , and determined RGS8 protein abundance by Western blot analysis ., In three independent experiments , one of which is shown in Fig 9C , levels of RGS8 decreased significantly in the presence of ATXN2-Q108 when compared with the levels in the presence of ATXN2-Q22 ., No significant alteration in the levels of RGS8 synthesis was detected between ATXN2-Q22 and control extracts ( Fig 9C and 9D ) ., These results suggest a role for ATXN2 in translational regulation and a dysregulation of this process in the presence of mutant ATXN2 ., Mouse models generated with tissue specific or strong promoters facilitate the evaluation of functional and anatomical consequences in many neurological disorders ., The Purkinje cell protein 2 ( Pcp2 ) and the Prion protein ( PrP ) promoters have been used to generate mouse models for polyQ ataxias such as SCA1 , SCA2 and SCA3 19 , 20 , 37–41 ., For instance , the use of the Pcp2 promoter for expressing mutant ATXN1 or ATXN2 has been shown to recapitulate the progressive cellular and functional phenotype of human SCA1 or SCA2 19 , 20 , 37 ., Use of a BAC-transgenic approach resulted in a more widespread expression of the transgene mirroring prior observations of endogenous ATXN2 expression in mouse and human 1 ., The control regions included in our BAC transgene specified expression in CNS and non-CNS tissues ( Fig 1B ) ., In the CNS , expression was seen in the cerebral hemispheres , cerebellum and spinal cord ., This is consistent with expression of endogenous mouse Atxn2 1 and in situ hybridization data as shown in the Allen Brain Atlas 28 ., In the cerebellum , expression of the BAC-transgene was seen in PCs , but also in granule cells , and neurons of the dentate nucleus ( Fig 2 ) ., As the transgenes were not tagged , we used LCM to establish transgene expression in these sub-regions of the cerebellum ., Future physiological experiments using the cerebellar slice preparation will need to examine what role mutant ATXN2 plays in granule cells and dentate nucleus and in overall cerebellar dysfunction in comparison with the PC-targeted expression of mutant ATXN2 20 ., Motor function deficits are common to all SCA2 mouse models , although their ages of onset differ ., The accelerating rotarod is used to measure motor coordination and motor learning over a number of days ., Our BAC-Q72 mice developed progressive motor deficits beginning at 16 weeks of age ( Fig 3B ) ., The motor phenotype of our BAC-Q72 mice was intermediate to that of our Pcp2-ATXN2Q58 and Pcp2-ATXN2Q127 mice , although transgene copy numbers and precise developmental expression patterns are difficult to compare ., As with our Pcp2-ATXN2Q22 line 19 , the BAC-Q22 line did not show a motor or cellular phenotype ., This study now extends these observations to mRNA measurements of key PC genes out to 45 weeks of age ( Fig 5A ) ., Lack of mRNA changes in BAC-Q22 are likely not due to differences in expression levels between lines , as transgenic ATXN2 had higher expression in the BAC-Q22 than in the BAC-Q72 line , both at the level of mRNA and protein ( Fig 1C and 1D ) ., Lack of any changes in genes that are typically altered early in Pcp2-ATXN2Q127 and BAC-Q72 supports the notion that simple overexpression of human wild-type ATXN2 does not cause significant PC pathology ., In contrast , motor function deficits in Atxn2-CAG42 knock-in mice were not evident until the age of 18 months 21 ., By comparing the motor functions in these four SCA2 transgenic mouse models , it is apparent that motor function deficits are dependent on CAG repeat length ., Consistent with this interpretation , knock-in Atxn1-CAG78 SCA1 mice developed neither ataxic behavior nor a neuropathological phenotype 42 , while knock-in Atxn1-CAG154 SCA1 mice did 43 ., Our BAC-Q72 transgenic mouse model , although generating lower levels of mutant ATXN2 expression in the cerebellum , develop motor deficits that resemble findings in human SCA2 patients ., These observations validate the notion that SCAs can be accurately modeled in mice ., Animal models for several polyQ diseases have shown alteration of body weight 21 , 43–45 ., In this study , BAC transgenic mice demonstrated reduced body weights ., The magnitude was similar to knock-in Atxn2-CAG42 mice and Atxn1-Q154/2Q mouse models 21 , 43 ., On the other hand , mice lacking Atxn2 exhibit obesity as a consequence of insulin resistance and altered lipid metabolism pathways 16 , 17 , 46 ., Increased weight loss due to reduced body fat has also been reported in other polyglutamine diseases , including Huntington disease 47 , 48 ., Of note , reductions in body weight were similar for BAC-Q22 and BAC-Q72 mice suggesting that with regard to the body weight phenotype a simple gain of function may be operative that is mirrored by obesity in loss of function models ., RGS proteins comprise a large family of more than 20 members that negatively modulate heterotrimeric G protein signaling ., They share a homologous RGS domain that functions to activate the GTPase of Gα proteins ., RGS8 is widely expressed in testis , brain , and cerebellar Purkinje cells 56 , 57 ., Mice lacking Rgs6 or Rgs9 exhibit motor function deficits and ataxia 32 , 33 ., Rgs8 knock-out mice were viable , fertile , and showed normal development , but have not been tested in detail for motor behaviors or PC morphology 57 ., Given the importance of a dysregulated mGluR1-ITPR1 axis in SCA2 pathology 13 , 58 , reduction in RGS proteins could further increase abnormally enhanced mGluR1 signaling ., We therefore examined RGS8 abundance in BAC-Q72 mice and Epstein-Barr virus immortalized human lymphoblastoid B ( LB ) -cells from SCA2 patients ( Fig 7 ) ., The results demonstrated that Rgs8 transcripts and protein abundance were significantly decreased in BAC-Q72 mice ( Fig 7A and 7B ) ., Consistent with this , SCA2-LB cells also demonstrated decreased RGS8 transcripts ( Fig 7C ) ., Next , we developed an in vitro model using SH-SY5Y cells ., Overexpression of mutant ATXN2 resulted in downregulation of RGS8 and this phenomenon was not seen for other known ATXN2 interactors ( Fig 8 ) ., As protein levels appeared somewhat depressed out of proportion to the observed reduction in steady-state mRNA levels , we hypothesized that ATXN2 might regulate translation of mRNAs directly ., Consistent with this hypothesis , we showed that both wild-type and mutant ATXN2 immunoprecipitated RGS8 mRNA in human cell culture and that this interaction was weaker for mutant ATXN2 ( Fig 9A and 9B ) ., This was also reflected in in vitro translation assays as presence of an expanded polyQ tract in ATXN2 reduced translation ( Fig 9C and 9D ) ., Our observations are consistent with studies of the Drosophila homolog of ATXN2 ( Atx2 ) ., Atx2 regulates PERIOD ( PER ) translation by interacting with TWENTY-FOUR ( TYF ) that is required for circadian locomotor behavior ., Depletion of Atx2 or expression of mutant Atx2 protein blocked the recruitment of PABP to the TYF-containing protein complex and decreased abundance of PER , thereby altering behavioral rhythms 59 , 60 ., ATXN2 interactions with polyA-binding protein 1 ( PABPC1 ) , the splicing factor A2BP1/FOX1 and poly-ribosomes further support roles for ATXN2 in RNA metabolism 5–7 ., Depletion of PABP from a cell free extract prevented initiation of mRNA translation 61 ., Our studies now extend these observations to mammalian systems and to a gene abundantly expressed in PCs ., It is quite likely that Rgs8 will be just one member of a larger set of mRNAs whose expression is regulated by ATXN2 ., Aberrant RNA metabolism including processing , degradation , and translation is now recognized to play an important role in neurodegenerative diseases ., Among these diseases are amyotrophic lateral sclerosis ( ALS ) , Spinal Muscular Atrophy ( SMA ) and Fragile X syndrome ( FXS ) 62–70 ., Although ATXN2 had been implicated in steps regulating mRNA translation and formation of stress granules 8 , 71 , 72 , to our knowledge we describe for the first time a significant difference in these functions between wild-type and mutant ATXN2 ., Our observations may also have implications for ALS as long normal ATXN2 alleles are a risk factor for ALS 18 , 73 and some individuals with full mutant ATXN2 alleles may present as ALS 74 ., In summary , BAC-SCA2 transgenic mice represent the first animal model with expression of mutant full-length human ATXN2 under the control of its endogenous human promoter including intronic regulatory sequences ., These sequences resulted in widespread expression of ATXN2 mirroring expression of endogenous Atxn2 ., Expression of mutant ATXN2-Q72 , but not wild-type ATXN2-Q22 , led to a progressive motor deficit , accompanied by morphological and transcriptome changes ., As previously demonstrated in C . elegans and the fly 6 , 59 , 60 , 75 , ATXN2 may exert translational control upon a subset of mRNAs ., We showed in two independently generated models that presence of mutant ATXN2 in vivo resulted in reduced steady-state levels of RGS8 mRNA and even further reduction in RGS8 protein ., ATXN2 coprecipitated with RGS8 mRNA and mutant ATXN2 reduced translation of RGS8 mRNA ., RGS proteins can act via Gαq on G-protein coupled receptors ., As mutant ATXN2 enhances Ca2+ release from the endoplasmic reticulum ( ER ) via its abnormal interaction with ITPR1 , reduction of RGS8 might be pred

Introduction, Results, Discussion, Materials and Methods

Spinocerebellar ataxia type 2 ( SCA2 ) is an autosomal dominant disorder with progressive degeneration of cerebellar Purkinje cells ( PCs ) and other neurons caused by expansion of a glutamine ( Q ) tract in the ATXN2 protein ., We generated BAC transgenic lines in which the full-length human ATXN2 gene was transcribed using its endogenous regulatory machinery ., Mice with the ATXN2 BAC transgene with an expanded CAG repeat ( BAC-Q72 ) developed a progressive cellular and motor phenotype , whereas BAC mice expressing wild-type human ATXN2 ( BAC-Q22 ) were indistinguishable from control mice ., Expression analysis of laser-capture microdissected ( LCM ) fractions and regional expression confirmed that the BAC transgene was expressed in PCs and in other neuronal groups such as granule cells ( GCs ) and neurons in deep cerebellar nuclei as well as in spinal cord ., Transcriptome analysis by deep RNA-sequencing revealed that BAC-Q72 mice had progressive changes in steady-state levels of specific mRNAs including Rgs8 , one of the earliest down-regulated transcripts in the Pcp2-ATXN2Q127 mouse line ., Consistent with LCM analysis , transcriptome changes analyzed by deep RNA-sequencing were not restricted to PCs , but were also seen in transcripts enriched in GCs such as Neurod1 ., BAC-Q72 , but not BAC-Q22 mice had reduced Rgs8 mRNA levels and even more severely reduced steady-state protein levels ., Using RNA immunoprecipitation we showed that ATXN2 interacted selectively with RGS8 mRNA ., This interaction was impaired when ATXN2 harbored an expanded polyglutamine ., Mutant ATXN2 also reduced RGS8 expression in an in vitro coupled translation assay when compared with equal expression of wild-type ATXN2-Q22 ., Reduced abundance of Rgs8 in Pcp2-ATXN2Q127 and BAC-Q72 mice supports our observations of a hyper-excitable mGluR1-ITPR1 signaling axis in SCA2 , as RGS proteins are linked to attenuating mGluR1 signaling .

Spinocerebellar ataxia type 2 ( SCA2 ) is an inherited neurodegenerative disorder leading to predominant loss of Purkinje cells in the cerebellum and impairment of motor coordination ., The mutation is expansion of a protein domain consisting of a stretch of glutamine amino acids ., We generated a mouse model of SCA2 containing the entire human normal or mutant ATXN2 gene using bacterial artificial chromosome ( BAC ) technology ., Mice expressing a BAC with 72 glutamines ( BAC-Q72 ) developed a progressive cerebellar degeneration and motor impairment in contrast to mice carrying the normal human gene ( BAC-Q22 ) ., We found that even prior to behavioral onset of disease , the abundance of specific messenger RNAs changed using deep RNA-sequencing ., One of the mRNAs with early and significant changes was Rgs8 ., Levels of Rgs8 protein were even further reduced than mRNA levels in BAC-Q72 cerebella suggesting to us that mutant ATXN2 might have a role in mRNA stability and translation ., Using a cellular model , we showed that the ATXN2 protein interacted with RGS8 mRNA and that this interaction differed between normal and mutant ATXN2 ., Presence of mutant ATXN2 resulted in reduced RGS8 protein translation in a cellular model ., Our studies describe a mouse model of SCA2 expressing the entire human ATXN2 gene and emphasize the role of ATXN2 in mRNA metabolism .

journal.pcbi.1000880

Computational Analysis of Phosphopeptide Binding to the Polo-Box Domain of the Mitotic Kinase PLK1 Using Molecular Dynamics Simulation

Mitotic cell division involves a tightly orchestrated series of events that precisely segregate an equal complement of chromosomes to two daughter cells ., Abnormalities in mitosis generate aneuploid cells containing an unequal distribution of chromosomes , which may represent a starting point for the genesis of cancer ., The polo-like kinase 1 ( PLK1 ) is an important of mitosis , working at different steps to facilitate mitotic entry , progression through the stages of chromosome segregation , and finally , mitotic exit 1–3 ., To do so , PLK1 must phosphorylate a wide range of protein substrates , yet operate in a manner that is tightly controlled in space and time 4 ., How these conflicting requirements for PLK1 activity are fulfilled during mitosis remains unclear ., However , recent findings suggest that PLK1 activity is frequently mis-regulated in human cancers ., Thus , PLK1 is overexpressed in a wide range of human tumours , with high expression levels often correlating with poor prognosis 5 ., PLK1 consists of two distinct functional domains: an N-terminal kinase domain responsible for catalytic activity , and a C-terminal polo-box domain ( PBD ) , which binds PLK1 target proteins ., A flexible linker of approximately 50 amino acids joins these two domains together ., The kinase activity leads to the phosphorylation and activation of a number of key mitotic proteins , notably Wee1 , CDC25c , BubR1 and CyclinB1 6–8 ., Studies have established that the PBD is a phosphopeptide binding domain which binds to the consensus phosphopeptide sequence Pro/Phe-Φ/Pro-Φ-Thr/Gln/His/Met-Ser-pThr/pSer-Pro/Φ , where Φ represents a hydrophobic residue 9–10 ., Elia et al also identified a high-affinity synthetic phosphopeptide for PLK1 , which has the sequence PMQSpTPL ., However , at the majority of positions in this sequence , there is no particular preference for specific residues ., This relatively broad specificity with respect to phosphopeptide binding allows PLK1 to bind a large set of phosphorylation-primed target proteins ., A comprehensive proteomic analysis identified 622 potential binding partners of PLK1 11 and at least 17 of these have been confirmed as binding partners 1 ., Structural elucidation of the PBD has shown that it adopts a unique fold which forms a narrow groove into which phosphopeptides bind 9 , 12–14 ., The structure of the PBD of PLK1 from the Protein Data Bank ( PDB ) with PDBID 3BZI 10 can be seen in Figure 1 ., The exact mechanism for the function of the PBD has not been definitively determined , but the evidence suggests that it provides a scaffold where proteins can bind after they have been phosphorylated at other sites by priming enzymes like cyclin-dependent kinases 10 ., It has been suggested that these bound proteins may then act as substrates for the kinase activity of PLK1 or may cause a conformational change allowing other substrates to bind to the kinase domain ., Such regulatory mechanisms are found in other kinases , where interaction domains such as FHA , SH2 , WW and 14-3-3 act as molecular switchboards , allowing interactions to occur with specific partners 15 ., The available crystal structures identify Trp414 , His538 and Lys540 as the key residues for phosphopeptide binding ., The importance of these residues has been confirmed by cell-based experiments where their mutation abolishes PBD binding capacity 10 , 16 ., Considering the phosphopeptide sequence , initial work highlighted a striking selectivity for serine at the −1 position and slight selectivity for proline at the +1 position , but very little selectivity at any other position 9 ., This lack of selectivity suggests that the phosphopeptide recognition site is highly promiscuous ., Whilst this is unusual , it is consistent with PLK1s multiple functions throughout mitosis ., However , the existence of such a large number of PBD-interacting phosphopeptides demands a molecular explanation ., One striking feature of the PBD crystal structures generated to date is the nature of the interfacial contacts made between the PBD and bound phosphopeptide ., Yun et al have recently crystallised a variety of short phosphopeptides complexes with the PBD 14 ., Peptides as small as HSpTP and LHSpT were shown to bind to the PBD ., This suggests that a large proportion of the binding energy is contributed by the core SpT motif ., This suggestion is supported by analysis of the two complexes of the PBD with PMQSpTPL from PDBID 1UMW and LLCSpTPN from PDBID 3BZI ., Both Gln and Cys residues form intermolecular interactions with the PBD at the −2 position and both Leu and Gln residues form intermolecular interactions at the +2 position 14 ., The ability of very different residues to make contacts further supports the idea that the core SpT motif is the main determinant of binding ., A second striking feature of the crystal structures of the PBD-phosphopeptide complexes is the large fraction of interactions between the phosphopeptide and the protein that are bridged by water molecules ., For example , in PDBID 3BZI , there are 18 hydrogen bonding interactions between the PBD and the CDC25c phosphopeptide , 9 of which are bridged through water molecules ., The number of interactions bridged by water molecules is significantly greater than in other phosphopeptide-binding proteins ., For example , the phosphopeptide bound with the 14-3-3 protein from PDBID 1YWT 17 makes 13 hydrogen bonding interactions at the interface but only 1 through a water molecule ., In other examples , the phosphopeptide bound with the SRC-SH2 domain from PDBID 2PIE 18 makes 16 hydrogen bonding interactions at the interface with only 4 through water molecules , and the phosphopeptide bound with the Rad53p-FHA1 domain from PDBID 1G6G makes 15 hydrogen bonding interactions , none of which are mediated by water molecules 19 ., Therefore , bridging water molecules appear to play a specific and important role in PBD-phosphopeptide interactions ., The overall aim of the study was to explore the energetics and dynamics of PLK1 PBD interactions and probe the nature of the water molecules at the interface and how they affect binding ., To assess the determinants of binding and gauge the role of water molecules in phosphopeptides binding to the PBD of PLK1 , we performed molecular dynamics ( MD ) simulations to study the motion of atoms within the complexes ., Crucially , this approach captures the dynamic aspects of interactions that are ignored by calculations performed on static systems 20 ., It has also proved to be more accurate than other methods for estimating binding free energies 21 ., The strengths of the interactions were estimated using a molecular mechanics , Poisson-Boltzmann surface area ( MM-PBSA ) approach 22–23 ., This approach has been used in the past for analysing the role of water molecules at protein-protein interfaces 24 ., The determinants of affinity and the importance of water molecules at peptide binding interfaces have also been explored using MD simulations , for the SRC SH2 domain 25 and at a small molecule binding interface for the GRB2 SH2 domain 26 ., In addition , we assessed binding energy contributions from discrete residues within the phosphopeptide chain in an attempt to provide an energetic framework which explains the ability of PBD to accommodate such a wide range of phosphopeptide ligands and , therefore , to function with such a diverse range of targets throughout mitosis ., Lastly , we employed inhomogeneous fluid solvation theory to predict the enthalpy and entropy of hydration sites on the surface of the apo protein and the phosphopeptide complex in order to understand the importance of water molecules in mediating the intermolecular interactions ., This understanding allows us to elucidate the mechanisms controlling affinity and specificity in this system , providing biologically relevant information and facilitating drug development efforts ., The protein structures were initially prepared as follows ., Atom coordinates for the protein , the phosphopeptide , and the water molecules were taken from the PDB ., The hydrogen-atom positions for the protein and the water molecules were then built using the HBUILD facility of the CHARMM ( version 34b1 ) program 27 with the CHARMM22 energy function 28 ., Histidine residues were checked for protonation state manually ., His382 was assigned as epsilon protonated and His 538 was assigned as positively charged for the phosphopeptide complex and as delta protonated for the peptide complex and the apo protein ., All remaining histidines were assigned as delta protonated ., The residues lysine , arginine , aspartate , glutamate , cysteine , and tyrosine were also analyzed to check their protonation state ., There was no evidence of any unusual protonation states and thus all lysine and arginine residues were assigned as positively charged , all aspartate and glutamate residues were assigned as negatively charged , and all cysteine and tyrosine residues were assigned as neutral ., The atomic charges of the standard residues were assigned from the CHARMM22 forcefield ., The phosphate moiety from the phosphothreonine residue was assigned to be doubly deprotonated , as this is likely to be the dominant species at physiological pH of around 7 . 0 , particularly when in close proximity to the positively charged binding site ., The atomic charges of the dianionic phosphothreonine residues were assigned from the CHARMM27 forcefield , which is based on the charges of methylphosphate 29 ., We performed MD simulations at 300 K on the apo state , the peptide complex and the phosphopeptide complex to investigate the dynamic nature of the interactions between the protein , the peptide or phosphopeptide and the water molecules ., All three structures were prepared separately using the process schematically represented in Figure 2 ., In the first stage of preparation , the system was solvated with TIP3P water molecules 30 around the binding site ., All the water molecules observed in the crystal structure were retained before solvating with the sphere of water molecules ., Any water molecule overlapping with the protein , the crystal structure water molecules , the peptide or the phosphopeptide were removed ., The solvent sphere of radius 20 Å was centred at the coordinates of the heavy atom centroid of the CDC25c phosphopeptide from 3BZI ., The sphere completely enclosed the peptide and the binding site residues , extending at least 5 . 0 Å from the peptide or phosphopeptide ., This assembly was partitioned into a 16 Å/20 Å reaction region/buffer region for stochastic boundary MD 27 ., The solvent was then minimized by steepest descent ( SD ) for 5000 steps and then subjected to a 5 ps Langevin dynamics equilibration period at 300 K , during which the solute atom positions were fixed ., The binding site was then repacked with TIP3P water molecules to fill any gaps ., Any water molecule overlapping with the protein , the existing water molecules , the peptide or the phosphopeptide was removed ., The solvent was again minimized by SD for 5000 steps and then subjected to another 5ps Langevin dynamics equilibration period at 300 K , during which the solute atom positions were fixed ., The entire binding site was then subjected to minimisation by SD to allow the solute to adjust to the solvent for 2000 steps with heavy atoms fixed , for 5000 steps with main chain atoms fixed and then 10000 steps with no atom positions fixed ., This was followed by a 10 ps Langevin dynamics equilibration period when the temperature was raised from 240 K to 300 K . Finally , the entire binding site was equilibrated at 300 K for 50 ps ., We ensured that the system was brought to equilibrium before beginning the MD simulation by verifying that the system reached a point where the energy fluctuations were stable ., Production simulations were then performed for 10 . 0 ns at 300 K . During all CHARMM dynamics simulation , the positions of the main-chain heavy atoms were restrained using a 5 . 0 kcal/mol/Å2 harmonic force and the positions of the sidechain heavy atoms were restrained using a 1 . 0 kcal/mol/Å2 harmonic force ., The MD simulations were performed using the CHARMM ( version 34b1 ) program 27 with the CHARMM22 force field 28 and using the SHAKE 31 algorithm to constrain the bonds to hydrogen , allowing an MD time step of 1 . 0fs ., The simulations were performed using a deformable boundary potential with a Langevin friction coefficient of 62 . 0 ps−1 applied to the water molecule oxygen atoms 32 ., Electrostatic interactions were modelled with a uniform dielectric and a dielectric constant of 1 . 0 throughout the setup and production runs ., This protocol has been used previously to analyse the dynamics of water molecules in the HSP90 system 33 ., To explore the dynamics in the unbound state , simulations were performed on the peptide and phosphopeptide structures alone ., In these cases the protein was deleted before preparing the system and the equilibration procedure in Figure 2 was employed ., For the MM-PBSA calculations , we calculated the difference in free energy between the protein-ligand complex and the unbound protein plus the unbound ligand ., MM-PBSA calculations were performed at intervals of 10 ps from each 10 . 0 ns run to yield 1000 snapshots ., All of the water molecules were deleted and so not included explicitly in any of the MM-PBSA terms ., For the electrostatics interactions a dielectric constant of 2 . 0 was employed , as this reflects the dielectric constant within the protein interior ., The free energy change upon binding was calculated using the following equation: ( 1 ) EMM is the molecular mechanics ( MM ) interaction energy between the receptor and the ligand , ΔGPB and ΔGSA are the electrostatic and non-polar contributions to desolvation upon ligand binding , respectively , and −TΔS is the conformational entropy change , Evdw is the van der Waals interaction energy , Eelec is the electrostatic interaction energy and ΔEdeformation is the difference in internal energy between the bound state ligand and the unbound state ligand ., This is termed the ligand deformation penalty ., To calculate the ligand deformation penalty , we ran two separate MD simulations of the unbound ligands ., MM-PBSA calculations were performed at intervals of 10 ps from each 2 . 0 ns run to yield 200 snapshots ., The MM energies ( ΔEMM ) were calculated in CHARMM 27 for each snapshot ., The deformation penalty of the ligand was considered by considering the MM energy of the peptide in both the bound state and unbound state simulations ., This comprised electrostatic , van der Waals , and torsional contributions ., We did not include the deformation penalty of the protein , as this involves taking the difference between the two large values of the protein internal energy ., Even small errors in the individual energetic terms of these absolute internal energies can have a large impact on the relatively small energy differences and thus introduce large errors to the predicted binding energy 39 ., The solvent accessible surface area ( SASA ) calculations were performed using CHARMM by calculating the change in surface area upon binding in Å2 multiplied by a constant value of 0 . 00542 kcal/mol plus the constant value of 0 . 92 kcal/mol ., To determine the key interactions between the phosphopeptide and the protein , we also calculated the per-residue contribution to the binding free energy for the phosphopeptide ., Only sidechain atom contributions were included for each residue and the contributions from the mainchain atoms were calculated separately ., The intramolecular interaction energies between each pair of residues was split in half and assigned evenly between the two residues ., The Poisson-Boltzmann ( PB ) desolvation penalties were calculated with only the specific residue being considered ., All other ligand atoms were deleted ., As the desolvation is not pairwise additive , the sum of each desolvation piece differs to the total calculated with the complete ligand ., The SASA terms were calculated for each residue without the SASA constant term ., For all simulations , we also considered the standard error of the mean by dividing the simulation into 20 blocks of equal time and calculating the standard deviation of the individual components of the binding free energy ., The PB calculations to determine ΔGPB were performed using a modified version of the DelPhi program 34–35 at a 129×129×129 grid resolution with focusing boundary conditions 36 ., The molecular surface was used to represent the dielectric boundary , a dielectric constant of 2 . 0 was used for the molecular interior and a dielectric constant of 80 . 0 was used in the solvent region ., An ionic strength of 0 . 145 M with a Stern layer of 2 . 0 Å was used for all PB calculations ., For each snapshot , separate calculations were performed for the complex , the unbound protein and the unbound peptide ., Protein atoms were assigned PARSE charges 37 before the DelPhi calculations ., The ΔGPB for the Poisson-Boltzmann portion of the free energy changes were then calculated with the following equation: ( 2 ) Gpb ( complex ) , Gpb ( protein ) and Gpb ( peptide ) are the PB solvation energies of the complex , the protein and the peptide respectively ., An estimate of the vibrational entropy change upon binding was performed , using normal mode analysis of the heavy atom fluctuations 38 ., We only included the entropy change for the ligand , as simulation did not include the entire protein ., Separate calculations were performed for the bound ligands and the unbound ligands and quasiharmonic analysis was used to estimate the vibrational entropies of the bound and unbound states ., The VIBRAN module of the CHARMM program 27 was used to determine normal modes and normal-mode frequencies by diagonalisation of the force constant matrices ., We used the entire 10 . 0 ns trajectories with a 20 . 0 fs timestep for the calculations ., Water molecules were not included in this analysis ., Translational and rotational motions were projected out from the dynamics trajectories by reorienting all the species using mass weighting ., The frequencies of the vibrational modes for the heavy atoms were then computed at 298 K using a quasiharmonic approximation ., The vibrational entropy of each system was then estimated from the vibrational frequencies 38 ., Further binding energy calculations were performed on a static structure as a comparison ., We began with the prepared phosphopeptide and peptide complexes ., To adjust the complexes for use with CHARMM 27 , the crystal structures were first subjected to a geometry optimization with the CHARMM22 energy function 28 ., All receptor sidechains were harmonically restrained with a force constant of 1 . 0 kcal/mol/Å2 and all receptor backbone atoms were fixed ., The minimization was performed for 1 , 000 , 000 steps using the adopted basis Newton-Raphson method ., All water molecules were then deleted and the van der Waals , electrostatic and SASA terms were then calculated with CHARMM ., To calculate the ligand deformation , the ligand structure was subjected to geometry optimization , separated from the receptor , for 1 , 000 , 000 steps ., It was necessary to place harmonic restraints with a force constant of 1 . 0 kcal/mol/Å2 on all atoms to prevent the ligands from collapsing ., The desolvation calculations were then performed using DelPhi 34–35 as described above ., This can be considered a single-point MM-PBSA calculation ., We performed additional analysis of the water molecules in the CHARMM MD simulations ., Initially , we calculated the mean enthalpy of water molecules at specific sites at the binding interface ., For each site , we analysed 1000 snapshots at 10 . 0 ps intervals across the 10 . 0 ns CHARMM simulation of the apo protein ., For each snapshot , we considered every water molecule within 1 . 4 Å of each point and summed the MM energy ., This allowed us to derive the mean value of the enthalpy for water molecules around each site ., The relative energy with respect to bulk water was calculated by subtracting the mean enthalpy of a bulk water molecule ., This was calculated using the MM energy from an MD simulation of a box of water molecules ., We performed a 2 . 0ns simulation of a sphere of water molecules with 16Å/20Å reaction region/buffer region for stochastic boundary MD ., To exclude water molecules at the surface of the sphere , only water molecules within 12 . 0 Å of the centre of the 20 . 0 Å sphere were included in the calculations ., The mean value of the enthalpy of a bulk water molecule was calculated using 200 snapshots and found to be −18 . 5 kcal/mol ., However , such an analysis only considers the enthalpic contribution to the free energy , ignoring the solvent entropy ., We thus performed further calculations to explore the solvent enthalpy and entropy across the entire binding interface ., There are a number of methods used to include the effects of solvent entropy on the free energy of water molecules ., These include free-energy perturbation 39–40 , thermodynamic integration 41 and inhomogeneous fluid solvation theory 42 ., We chose to employ inhomogeneous fluid solvation theory , first described by Lazaridis 42 and implemented in Schrödingers WaterMap software 43–45 ., In this method , molecular dynamics simulations are analysed to cluster distinct hydration sites and assign an enthalpy and entropy to each one ., The enthalpy is calculated as the average interaction energy over the simulation ., The entropy is calculated by comparing the distributions of translations and orientations available to the water molecule in bulk water and at the surface ., We performed calculations on the apo state and the phosphopeptide complex ., The WaterMap MD simulations were run with Desmond 46 using the OPLS_2005 force field 47 ., We began with the prepared phosphopeptide complex and the apo protein ., All water molecules from the crystal structure were deleted and TIP4P water molecules 48 were added with the System Builder module in Maestro ., The solvated structure was then subjected to restrained minimisation using a force constant of 5 . 0 kcal/mol/Å2 on the solute heavy atoms ., This was followed by a molecular dynamics simulation of 48 . 0 ps in which the temperature of the system was increased from 10 to 300 K . The harmonic restraints of 5 . 0 kcal/mol/Å2 on solute heavy atoms were retained ., A preproduction simulation was then run at 300 K for 120 . 0 ps ., Finally , the production simulations were run for 2 . 0 ns in the NPT ensemble at a temperature of 300 K and a pressure of 1 atm ., The statistical analysis was performed using snapshots from the production simulation ., Water molecules in the proximity of the binding site from 2000 equally spaced snapshots were clustered to form hydration sites ., For each hydration site , the enthalpy was computed as the average non-bonded energy of each water molecule within the hydration site with the rest of the system ., The excess entropy was computed by numerically integrating a local expansion of spatial and orientational correlation functions 44 ., Only contributions from the first-order term of the expansion were included ., At this stage , we have not incorporated the effect of the solvent free energy on predictions of the total binding free energy of the phosphopeptide ., A complete treatment of the energetics of the solvent would involve a consideration of all water molecules in the complex and the apo protein ., It would also require a consideration of the unbound ligand ., However , no prediction of the binding free energy is complete without such consideration ., Protein surfaces can generate highly hydrophobic regions and this creates volumes of space where water molecules have unfavorable free energies ., Filling these volumes with hydrophobic ligand atoms is a general mechanism to increase the ligand binding affinity ., In some cases , there is a suggestion that water molecules will completely evacuate extremely hydrophobic cavities 43 ., The creation of a small vacuum region leads to a large energetic penalty and filling such a cavity with a small molecule should greatly increase its affinity ., To test the predictions made by the MM-PBSA calculations , the affinities of the CDC25c peptide and phosphopeptide were measured experimentally by fluorescence polarization ( FP ) 49 and by isothermal titration calorimetry ( ITC ) ., Human PLK1 amino acids 345 to 603 were amplified by PCR and cloned into the EcoRI and NotI sites of the bacterial expression vector pGEX6P-1 , expressed in E . Coli and purified as previously described 12 ., All peptides and phosphopeptides were synthesised using standard chemistry ( Designer Bioscience Ltd . , Cambridge , UK ) ., The fluorescently labelled probe was the phosphopeptide sequence MAGPMQSpTPLNGAKK with N-terminal TAMRA ., The peptide competitor was the sequence LLCSTPNGL and the phosphopeptide competitor was the sequence LLCSpTPNGL ., FP measurements were carried out in a 384-well , low-volume , black , flat bottom polystyrene NBS microplate ( Corning 3820 ) using a PHERAstar Plus plate reader ( BMGLabtech ) ., The final reaction volume of 45 µl contained 10nM labelled probe peptide , 35nM PBD and varying concentrations of competitor ., FP values were obtained in millipolarisation units at an excitation wavelength of 540 nm and an emission wavelength of 590 nm , and were calculated in terms of percentage inhibitions ., ITC measurements were performed using a VP-ITC microcalorimeter ( MicroCal Inc . ) ., The experiments consisted of injecting CDC25c peptide or CDC25c phosphopeptide at a concentration of 120 µM into a sample cell containing 12 µM of PBD in 50mM Hepes pH 7 . 4 , 200mM NaCl , 1 mM EDTA , 1mM EGTA ., Fifty injection of 4 . 5 µl of LLCSpTPNGL were performed with a spacing of 180 s using a reference power of 25mCal/s ., Thirty injections of 8 µl of LLCSTPNGL were performed with a spacing of 240 s using reference power of 25mCal/s ., All binding isotherms were analysed and graphed using Origin Software 7 . 5 ( MicroCal Inc . ) ., The first calculation we performed was the prediction of the MM-PBSA binding free energy change for both the peptide and the phosphopeptide ., The predicted binding free energy changes calculated by this method are presented in Table, 1 . The CDC25c phosphopeptide is predicted to bind with greater affinity than the corresponding peptide due to the more favourable binding enthalpy ., As expected , the entropy changes of both ligands are unfavourable , due to the restriction on the ligand poses in the bound state ., The calculations performed on the static structure are also presented in Table, 1 . Both the enthalpy and the change in enthalpy from the static calculations are similar to those predicted by the dynamic MM-PBSA calculations ., This suggests that for this system , static calculations may be sufficient to predict relative binding enthalpies ., However , such static calculations ignore the change in both the solute and the solvent entropy that we aimed to quantify with the MD simulations ., To estimate the importance of each residue in the phosphopeptide , we analysed the per-residue contributions to the MM-PBSA binding enthalpy ., The results are presented in Table, 2 . They highlight the importance of the phosphothreonine residue , which contributes over 30% of the binding enthalpy of the phosphopeptide ., In fact , the phosphothreonine residue and the mainchain atoms together contribute over 75% of the binding enthalpy of the phosphopeptide ., This begins to explains why the PBD does not discriminate strongly between different phosphopeptides , as in this case only a small contribution to the binding enthalpy is made by the non-phosphothreonine sidechains ., The residue Leu1 makes a reasonable contribution to binding due to its strong van der Waals interactions with a hydrophobic surface ., This is consistent with the experimental data from oriented peptide library screening , as leucine is one of the residues selected for at the −4 position 10 , 14 ., Residues Cys3 and Asn7 are predicted to make a small unfavourable contribution to binding ., This is also consistent with the experimental data , as these residues are not selected at the −2 and +2 positions ., However , residues Leu2 and Ser4 are predicted to make very little contribution to the binding enthalpy , but are selected at positions −3 and −1 respectively 9–10 ., The importance of these residues is revealed by considering the interactions of the solvent in the later sections ., The MM-PBSA calculations predict that the CDC25c phosphopeptide will bind to the PBD with higher affinity than the CDC25c peptide ., We tested this prediction experimentally using an FP assay and by ITC ., We measured the effect of both the peptide and the phosphopeptide on binding of a fluorescently tagged phosphopeptide ., The FP results can be seen in Figure 3 and show that , as predicted , the CDC25c phosphopeptide has a significantly higher affinity than the CDC25c peptide , which showed no measurable binding ., We also measured the binding of the CDC25c peptide and the CDC25c phosphopeptide by ITC ., The ITC data can be seen in Figure 4 and confirms that the CDC25c peptide shows no detectable binding and the CDC25c phosphopeptide binds with a measured affinity of 0 . 705 µM ., These results are consistent with the biological function of PLK1 and with prior experimental work on other phosphopeptides , but have not previously been explained quantitatively ., In order to understand the dynamic nature of the system , we looked at the trajectory of each individual water molecule in each simulation and calculated the RMSF of the oxygen atom from its mean position ., The water molecules in the sphere of water molecules are highly mobile , with a mean RMSF of 11 . 5 Å ., No water molecule has an RMSF below 2 . 0 Å ., However , in all three protein simulations there are a large number of water molecules near the protein surface with an RMSF below 0 . 5 Å ., It is clear that even in the apo state , water molecules are fixed to some degree at the surface ., It is possible that a larger sphere of water molecules or a longer timescale is needed to accurately model the mobility of these surface water molecules ., However , the conclusion is supported by analysis of the water molecules in the apo crystal structures , which have B factors similar to the protein residues ., A closer examination of individual water molecules across the simulation reveals the presence of distinct hydration sites at the surface , formed by hydrogen bonding interactions with the protein ., Figure 5a shows four such hydration sites , occupied by one water molecule during the course of the 10 . 0ns simulation of the apo protein ., Water molecules at the circled site form a hydrogen bond to the backbone amide of residue Trp414 ., Water molecules in this hydration site are expelled upon ligand binding and replaced by the serine residue at the −2 position of the consensus sequence ., We estimated the enthalpy of water molecules at this site with respect to bulk water for each of the 1000 snapshots of the apo protein ., We placed the hydration site at a position 3 . 0 Å along the amide nitrogen to amide hydrogen bond vector ., For each snapshot , we considered every water molecule within 1 . 4 Å of the site , representing the idealized radius of a water molecule ., There is a water molecule within this sphere in 99% of snapshots , whilst the average enthalpy of water molecules within the sites with respect to bulk water is +1 . 3±1 . 5 kcal/mol ., Expulsion of water molecules from this site should thus provide an enthalpic bonus upon ligand binding ., However , water molecules at the circled hydration site show marked translational and orientational ordering , as shown in Figur

Introduction, Materials and Methods, Results, Discussion

The Polo-Like Kinase 1 ( PLK1 ) acts as a central regulator of mitosis and is over-expressed in a wide range of human tumours where high levels of expression correlate with a poor prognosis ., PLK1 comprises two structural elements , a kinase domain and a polo-box domain ( PBD ) ., The PBD binds phosphorylated substrates to control substrate phosphorylation by the kinase domain ., Although the PBD preferentially binds to phosphopeptides , it has a relatively broad sequence specificity in comparison with other phosphopeptide binding domains ., We analysed the molecular determinants of recognition by performing molecular dynamics simulations of the PBD with one of its natural substrates , CDC25c ., Predicted binding free energies were calculated using a molecular mechanics , Poisson-Boltzmann surface area approach ., We calculated the per-residue contributions to the binding free energy change , showing that the phosphothreonine residue and the mainchain account for the vast majority of the interaction energy ., This explains the very broad sequence specificity with respect to other sidechain residues ., Finally , we considered the key role of bridging water molecules at the binding interface ., We employed inhomogeneous fluid solvation theory to consider the free energy of water molecules on the protein surface with respect to bulk water molecules ., Such an analysis highlights binding hotspots created by elimination of water molecules from hydrophobic surfaces ., It also predicts that a number of water molecules are stabilized by the presence of the charged phosphate group , and that this will have a significant effect on the binding affinity ., Our findings suggest a molecular rationale for the promiscuous binding of the PBD and highlight a role for bridging water molecules at the interface ., We expect that this method of analysis will be very useful for probing other protein surfaces to identify binding hotspots for natural binding partners and small molecule inhibitors .

Cell division is a key biological process and imperfections in the process can lead to diseases like cancer ., Polo-Like Kinase 1 ( PLK1 ) is a protein kinase enzyme that controls cell division by interacting with many other proteins ., Malfunction of PLK1 has been implicated in cancer ., To understand how PLK1 interacts with so many other proteins , we created a three-dimensional model of PLK1 and simulated its dynamic nature ., Analysis of the components of the binding affinity provided insight into how the binding specificity is achieved ., We also employed a method of analysis that locates regions of the protein surface that are particularly important in controlling binding affinity ., Our results not only provide a valuable tool that can be generally applied to analyzing the binding between protein surfaces , but also provide insights into how PLK1 works to control cell division by binding to specific partners ., In the future , these analyses could help to design drugs that block the interaction between PLK1 and its partners to block cell division for the treatment of diseases like cancer .

oncology, biochemistry/experimental biophysical methods, biochemistry/biomacromolecule-ligand interactions, biochemistry/macromolecular assemblies and machines, biochemistry/theory and simulation, computational biology/molecular dynamics, biochemistry/drug discovery

journal.pntd.0004833

Estimating Dengue Transmission Intensity from Case-Notification Data from Multiple Countries

Dengue is the most widely distributed mosquito-borne viral infection , but assessment of its geographic variation in transmission remains challenging ., Analysis based on mapping the probability of occurrence of dengue estimated that dengue causes 390 million annual infections worldwide 1 ., However , these estimates relied on assuming a direct linear correlation between the probability of occurrence and incidence , rather than estimating transmission intensity as quantified by the force of infection or reproduction number ., Here we develop methods to estimate transmission intensity from routine , age-stratified surveillance data on suspected dengue case incidence ., All four serotypes of dengue virus ( DENV-1 , 2 , 3 , and 4 ) can cause dengue fever with the risk of severe dengue increasing with subsequent heterologous infections ., Once infected , individuals develop long-lived protective homotypic immunity and short-lived heterotypic immunity 2 , 3 ., Once antibody levels wane below the threshold required to provide protection , antibody-dependent enhancement ( ADE ) becomes a risk , leading to secondary heterologous infection having an increased risk of causing clinically apparent disease 4 , 5 ., Hence , while the majority of primary dengue infections are asymptomatic 6 , 7 , secondary heterologous infection has been identified as a major risk factor for symptomatic and severe dengue 8–10 ., Therefore the majority of cases seen in hospitals 11 or reported via surveillance systems 12 tend to be secondary infections 7 ., In previous work , we estimated dengue transmission intensity from age-stratified seroprevalence data but highlighted the relative paucity of seroprevalence data compared with routine surveillance data on the incidence of suspected dengue 13 ., This reflects dengue fever , dengue haemorrhagic fever ( DHF ) , and dengue shock syndrome ( DSS ) being notifiable diseases in most countries 14–18 ., Indeed , in many countries , incidence reports are the only type of data available ., However the clinical diagnostic criteria vary and different countries have their own reporting standards 19 ., The World Health Organisation ( WHO ) collates surveillance data from dengue affected countries via its DengueNet system , but the data are not always updated regularly and there can be inconsistencies with other sources ( e . g . WHO regional offices or countries ) of national and subnational data 19 ., The lack of systematic data on dengue incidence , the lack of standardised reporting procedures or diagnostic criteria , and the lack of integration between private and public sectors makes accurate estimation of the true dengue burden difficult 20 ., Previous studies have attempted to estimate the burden of dengue and associated economic costs in South East Asia and South America by calculating expansion factors from systematic literature reviews , collation of existing data , and population-based cohorts 20–24 ., However , the lack of standardisation also affects the validity of expansion factors ( calculated by dividing the cumulative incidence of dengue cohort studies by that from passive data at national and local levels ) as estimates of underreporting ., Due to the wide spectrum of clinical manifestations and the lack of routine laboratory testing , dengue is globally underreported and analyses of officially reported dengue numbers need to take this into account 25 ., While reported incidence levels cannot be relied upon to directly quantify disease burden , the age distribution of dengue cases provides more reliable information on dengue transmission intensity ., Here we propose an approach for estimating average transmission intensity—as quantified by the force of infection ( λ ) or basic reproduction number ( R0 ) –from age-stratified incidence data ., We compare estimates derived from seroprevalence and incidence data and assess the level of under-reporting of dengue disease ., In addition , we estimate the relative contribution of primary to quaternary infections to the observed burden of dengue disease incidence ., Web of Knowledge and PubMed were searched for age-stratified incidence data since 1980 as we were interested in contemporary dengue transmission and wanted to be consistent with our previous study where we collated age-stratified seroprevalence data 13 ., Search terms used were ‘dengue’ and ‘age’ and ( ‘incidence’ or ‘cases’ or ‘notifications’ or ‘notified cases’ ) with inclusion criteria mapped to subject headings ., Additional web-based searches were performed to augment the primary literature search ., Data were extracted from published datasets where authors reported age-stratified incidence data with corresponding population age-structure data ., We considered a population stratified into M age groups and denote aj and aj+1 the lower and upper age bounds respectively of age group j ( j = 0 , … , M-1 ) ., Our model assumes perfect homotypic protection following infection with any serotype ., Thus , an individual can experience a maximum of four dengue infections in their life ( corresponding to the four dengue serotypes ) ., Ideally , we would allow forces of infection to vary by serotype ( DENV-1 to DENV-4 ) ., However as serotype-specific data were not available , we assumed circulating serotypes were equally transmissible , i . e . had the same force of infection , λ , which did not vary over time ., The incidence of primary infections ( I1 ) for any one serotype for people in an age group j was calculated as the integral of the probability of being seronegative to all four strains at age a multiplied by four times the constant serotype-specific infection hazard , λ ( since primary infection can occur with any of the four serotypes ) ., Age a spans the range aj , aj+1 , as described by the bounds of integrations ( Eq 1 ) ., The incidence of secondary , tertiary , and quaternary infections in age group j ( I2 ( j ) , I3 ( j ) , and I4 ( j ) respectively ) are calculated in a similar fashion ., If fewer than four serotypes have circulated in an area , then the number of infections an individual can have changes accordingly ., Full details are given in the Supporting Information ( S1 Text ) ., The average observed annual disease incidence rate per person in age group j is then given by the weighted sum of the primary to quaternary infection rates ( Eq 2 ) :, D ( j ) =ρw ( j ) {I2 ( j ) +γ1 ( I1 ( j ) +γ3 ( I3 ( j ) +I4 ( j ) ) ) +B}, ( 2 ), where w ( j ) = aj+1 − aj is the width of age group j , ρ is the probability that a secondary infection results in a detected dengue case ( reporting rate ) , γ1 is the probability that a primary infection is detected relative to a secondary infection , and γ3 is the probability that a tertiary or quaternary infection is detected relative to a primary infection ., Here B is a baseline risk of disease used to represent any non-dengue related illnesses that are misdiagnosed as dengue , and was only estimated when fitting suspected dengue incidence data where laboratory confirmation was lacking ., We assumed that secondary infections were more likely to be symptomatic than primary infections 7 , 26 and that post-secondary infections were even less likely to be symptomatic than primary infections , i . e . ρ>γ1>γ3 ., Single values of γ1 and γ3 were estimated per country ., For datasets that reported DHF only , we assumed that DHF cases only arose from secondary infections and set γ1 and γ3 to zero 27 , 28 ., Where fewer than four serotypes were in circulation , we adjusted our calculation of the expected incidence accordingly—full details are given in the S1 Text ., Where data on the age distribution of the population was not provided in the source publications , the population age-structure closest to the survey population was used ( taken from census data or from United Nations estimates ) 29 ., For the first model variant examined ( model 1 ) , we assumed a single baseline reporting rate ( ρ ) across all age groups ., We also explored whether baseline reporting rates might differ with age ( model 2 ) by estimating different reporting rates in children ( ρyoung ) and adults ( ρold ) , also fitting the age threshold ( athreshold ) defining the boundary between these groups ( ρyoung ) for age a < athreshold , otherwise ρold ) ., Where incidence data were available for multiple years , we fitted models 1 and 2 to individual years ( model variants 1A and 2A ) ., We also examined fitting to the cumulative incidence across the observation period , as this gives a better estimate of the long-term average distribution of incidence across age groups ( models 1B and 2B ) ., When fitting to the cumulative incidence we calculated the expected disease incidence by multiplying the annual expected disease incidence by the number of years in the study ., Overall , for models 1A and 1B , we estimated up to 5 parameters ( λ , ρ , γ1 , γ3 and B ) , while for models 2A and 2B we estimated up 7 parameters ( λ , ρyoung , ρold , athreshold , γ1 , γ3 and B ) ., All models were fitted to the data using a Metropolis-Hastings Markov Chain Monte Carlo ( MH MCMC ) algorithm using a Dirichlet-multinomial log-likelihood with uniform priors in version 3 . 1 . 0 of the R statistical language 30 ., Full details are given in the S1 Text ., We assumed dengue transmission was at endemic equilibrium and that the force of infection ( λ ) was constant in time ., Since we did not have serotype-specific data , we additionally assumed that all serotypes in circulation were equally abundant and equally transmissible , i . e . had the same force of infection and basic reproduction number , and that there were no interactions between serotypes ., We estimated a strain-specific basic reproduction number ( R0 ) from the single force of infection ( λ ) estimated under two different assumptions about the number of infections required to acquire complete immunity ., Under assumption one , complete protection is acquired upon quaternary infection ., Under assumption two , complete protection is reached after secondary infection ( or if tertiary and quaternary infections occur , they are not infectious ) ., These assumptions match that of our previous work estimating the force of infection from serological data and allowed us to compare the R0 estimates obtained from both types of data 13 ., Full details are given in the S1 Text ., We used weighted regression to assess how comparable force of infection estimates obtained from cumulative incidence data were with those derived from seroprevalence data described previously 13 and from four additional seroprevalence datasets ( see Table S1 in S1 Text ) ., Location- and time-matched incidence and serology data were not available , so we matched datasets by country , region , and survey year ., Since seroprevalence data represent all past infections , we compared force of infection estimates with those obtained from cumulative incidence data rather than yearly incidence data where possible ( see Table S2 in S1 Text for full details on pairings ) ., We used the weighted regression method described by Ripley and Thompson 31 which explicitly accounts for measurement errors in both force of infection estimates from seroprevalence data ( y-axis ) and incidence data ( x-axis ) to estimate the maximum likelihood estimate ( MLE ) line ., This was implemented using the deming package in R 32 ., Full details are given in the S1 Text ., We identified 23 papers reporting incidence data ., Fig 1 describes the search process and Table 1 summarises the studies identified ., Seven papers reported age-stratified incidence data from multiple years , one paper reported data where the number of serotypes in circulation had changed over the survey years , six papers reported cumulative age-stratified incidence data , eight papers reported age-stratified incidence data from a single year , and two papers reported age-stratified incidence data from multiple countries ., The identified studies provided a total of 34 datasets from 13 countries ., The years included ranged from 1978 to 2011 ., The dataset reporting incidence data from 1978 was included since data were presented for the eleven-year time period of 1978–1988 33 ., Of the 23 papers reporting incidence data , ten reported dengue incidence at the national level and only two studies reported cases detected via active as well as passive surveillance ., Three additional surveys were obtained from the Ministry of Health in Thailand that reported age-specific incidence from Bangkok ( 2000 ) , Ratchaburi ( 2000 ) , and Rayong ( 2010 ) 34 ., As expected , force of infection estimates varied widely between countries , with less variation seen within countries ., Fig 2 shows the distribution of the total force of infection ( λtotal ) grouped by country ( calculated by multiplying the serotype-specific force of infection by the number of serotypes in circulation ) ., Individual estimates are given in the S1 Text ., Estimates of R0 varied according to the assumptions made regarding host immunity ., Assuming only primary and secondary infections are infectious ( assumption two ) gave up to two-fold higher estimates of R0 than when assuming tertiary and quaternary infections are also infectious ( Fig 2 ) ., This is consistent with our previous results analysing seroprevalence data 13 ., Some force of infection estimates in Cambodia were very high , perhaps as a result of the active surveillance undertaken as part of the study by Vong et al . 38 ( for all parameter estimates see S1 Text ) ., The baseline reporting rate ( ρ ) , defined as the probability of detecting a secondary infection , was less than 15% when averaged across all studies ( Fig 3 ) ., The median probability of detecting a primary infection relative to that of detecting a secondary infection ( γ1 ) was less than 25% for the majority of datasets ., However , the credible intervals for some γ1 estimates were wide ., The data proved uninformative about the contribution of post-secondary infections to disease incidence , as our estimates of γ3 ( Fig 3 ) reflected the prior distribution assumed for that parameter ( uniform from 0 to 1 ) ., The baseline reporting rates ( ρ ) varied substantially by country ( Fig 3 ) , likely reflecting differences in healthcare seeking behaviour and surveillance ., Generally , estimated reporting rates in the Americas were higher than in South East Asia , with Singapore having the highest rate within SE Asia ., Reporting rates also varied within each country depending on survey year or survey region , which may reflect differences in local healthcare systems or changes in public awareness after epidemics ., We used weighted regression to compare the force of infection estimates obtained from age-stratified seroprevalence data to cumulative incidence data ., Estimates obtained from the model fitted to the cumulative incidence data were largely comparable to force of infection estimates from seroprevalence data ( Fig 4 ) ., The majority of the total force of infection ( λtotal ) estimates from incidence data ( calculated by multiplying the serotype-specific force of infection by the number of serotypes in circulation ) were comparable to those obtained from seroprevalence data when λtotal was smaller than ~0 . 1 with greater uncertainty as the force of infection increased ., In two of the three locations in Thailand where region and time matching seroprevalence and incidence data were available 34 , the force of infection estimates obtained from the models fitted to incidence data and serology data had overlapping 95% credible intervals ., In Ratchaburi the estimate obtained from seroprevalence data was smaller than that from incidence data ( Fig 5 ) ., From a literature search we selected 23 papers reporting age-stratified case notification data in 13 countries from 1978–2010 ., For each dataset we estimated dengue transmission intensity as quantified by the force of infection ( λ ) and the basic reproduction number ( R0 ) ., Where possible we fitted to the cumulative incidence data as fitting to yearly incidence data gave less stable estimates ( model fits to yearly incidence data are given in the S1 Text ) The total force of infection ( λtotal ) estimated from cumulative incidence data were then compared with previous λ estimates from seroprevalence data ., The incidence model presented in this paper provides a method for estimating dengue transmission intensity in areas where seroprevalence data are not available ., Force of infection estimates and corresponding basic reproduction numbers varied widely across and within countries as expected , highlighting the heterogeneous nature of dengue transmission spatially and temporally ., The majority of our R0 estimates ranged from 1 to 5 , similar to our estimates obtained from seroprevalence data 13 ., Similarly to our serology-based estimates , force of infection estimates were generally higher in South East Asia than for Latin America ., Since we had no serotype-specific notification data , we assumed that all serotypes were equally transmissible and equally abundant ., If serotype-specific notification data were available , serotype-specific forces of infection could be estimated ., Although we assumed that dengue transmission intensity does not vary with age , is constant in time and equal for all serotypes in circulation , previous studies have shown that transmissibility can differ substantially not only between serotypes 13 , 54 but also seasonally , yearly 54 , and spatially 55 ., However , given the available data it was not possible to estimate serotype-specific or time-varying forces of infection ., Multiple cross-sectional surveys or cohort studies are required to estimate how forces of infection have changed by age over time , and serotype-specific data are needed to resolve differences between serotypes ., Due to the lack of incidence and serology data collected in the same year and region , we matched cumulative incidence and serology datasets according to the year or region ( see S1 Text ) ., While overall estimates from incidence data were comparable with those derived from seroprevalence data , it would nonetheless be beneficial to validate this model with more incidence and serology datasets collected simultaneously in the same geographical location ., Generally , estimated reporting rates ( ρ ) in the Americas were higher than those in South East Asia with Singapore having the highest rate within South East Asia , consistent with their well-established dengue surveillance program 56 ., Reporting rate estimates also varied within each country depending on survey year or survey region reflecting variation in healthcare and surveillance systems 19 ., Reporting rates are also likely to change in response to recent or current epidemics which affect public awareness of dengue and thus healthcare seeking behaviour 57 ., Additionally , in an epidemic year clinicians may preferentially diagnose a febrile illness as dengue without laboratory testing 58 ., We hypothesised that severity or disease reporting differed by age group and estimated age-dependent reporting rates ( ρyoung and ρold ) and the age at which reporting rates changed ( Athreshold ) ., However due to the wide age bands of the available data , we were not able to explore this fully ., Full details are given in the S1 Text ., Since the majority of notified dengue cases are diagnosed as secondary dengue infections 4 , 5 , 7 , 11 , 12 , 59 , we assumed that the probability of detecting a primary case would be smaller than the probability of detecting a secondary case , and that the probability of detecting a tertiary or quaternary case would be smaller than the probability of detecting a primary case ( γ3<γ1<ρ ) ., The probability of detecting a primary case was consistently low relative to a secondary case ( Fig 3 ) at less than 50% , the majority being under 25% ., However , we were not able to estimate the probability of detecting a tertiary/quaternary case ( relative to a primary case ) from the available data ., A prospective cohort study in Nicaragua found that the proportion of inapparent to symptomatic infection did not differ according to whether an individual had a primary , secondary , or tertiary infection 60 ., Overall , the impact of cross-immunity and the contribution of tertiary and quaternary infections to onward transmission are not well quantified ., While there is evidence that tertiary and quaternary infections occur 61 , 54 , there is little quantitative data on the infectiousness or severity of such infections relative to primary and secondary infections ., Additionally , clinically apparent tertiary or quaternary infections are not routinely reported , nor can they be tested for retrospectively 61 ., Wikramaratna et al . showed that tertiary and quaternary infections allows for the high seroprevalence at very young ages observed in Haiti 62 and Nicaragua 63 better than when assuming complete protection after two heterologous infections 61 ., Since the majority of dengue infections are mild or asymptomatic , even sensitive healthcare systems can substantially underestimate true rates of infection even for the supposedly more severe secondary infections , as shown by the low baseline reporting rates 11 , 3 ., Furthermore , dengue has a wide spectrum of clinical manifestations making it difficult to accurately diagnose in the first instance 20 ., Our estimates from Thailand ( Fig 5 ) shows that even with data from the same location and year , it is difficult to make reliable comparisons between estimates obtained from seroprevalence and incidence data ., We were also comparing force of infection estimates from seroprevalence data to those from incidence data from a single year ( rather than cumulative incidence ) , which may have contributed to the observed discrepancy ., Although incidence data are the most abundant form of data available on dengue transmission , surveillance systems and reporting procedures are not standardised within or across countries making it very difficult to reliably compare estimates 20 ., Laboratory capacity and general public health infrastructure and surveillance systems vary widely and there is often no integration between private and public health sectors ., With such variable data , it is very difficult to estimate dengue burden ( or transmission intensity ) consistently ., Since non-serotype specific serological ( IgG ) surveys are relatively inexpensive to collect , it would be beneficial for such seroprevalence data to be collected routinely ., Such data would provide better baseline estimates of overall transmission intensity against which incidence based-estimates could be calibrated to assess changes in transmission and identify weaknesses in surveillance systems .

Introduction, Methods, Results, Discussion

Despite being the most widely distributed mosquito-borne viral infection , estimates of dengue transmission intensity and associated burden remain ambiguous ., With advances in the development of novel control measures , obtaining robust estimates of average dengue transmission intensity is key for assessing the burden of disease and the likely impact of interventions ., We estimated the force of infection ( λ ) and corresponding basic reproduction numbers ( R0 ) by fitting catalytic models to age-stratified incidence data identified from the literature ., We compared estimates derived from incidence and seroprevalence data and assessed the level of under-reporting of dengue disease ., In addition , we estimated the relative contribution of primary to quaternary infections to the observed burden of dengue disease incidence ., The majority of R0 estimates ranged from one to five and the force of infection estimates from incidence data were consistent with those previously estimated from seroprevalence data ., The baseline reporting rate ( or the probability of detecting a secondary infection ) was generally low ( <25% ) and varied within and between countries ., As expected , estimates varied widely across and within countries , highlighting the spatio-temporally heterogeneous nature of dengue transmission ., Although seroprevalence data provide the maximum information , the incidence models presented in this paper provide a method for estimating dengue transmission intensity from age-stratified incidence data , which will be an important consideration in areas where seroprevalence data are not available .

With 40% of the world’s population at risk of infection , dengue imposes a significant public health burden ., Yet estimates of baseline transmission intensity are still sparse , making it difficult to implement efficient control programs ., The authors used incidence data , which are abundant compared to seroprevalence data , to estimate dengue transmission intensity in 13 countries ., Estimates derived from incidence data were comparable to those from seroprevalence data , an important conclusion for areas where seroprevalence data are not available ., Additionally , the estimated baseline reporting rates and the contribution of primary to tertiary/quaternary infections to observed disease in each country will help to highlight potential weaknesses in the country or region’s surveillance system .

medicine and health sciences, pathology and laboratory medicine, infectious disease epidemiology, tropical diseases, geographical locations, age groups, neglected tropical diseases, infectious disease control, public and occupational health, infectious diseases, serology, epidemiology, dengue fever, people and places, infectious disease surveillance, asia, disease surveillance, population groupings, viral diseases, thailand

journal.pcbi.1000421

Maximum-Likelihood Model Averaging To Profile Clustering of Site Types across Discrete Linear Sequences

Analysis of discrete linear sequences has played an increasingly important role in biology ., In particular , the detection of heterogeneous regions among sequences can aid in understanding the heterogeneous processes that act upon those regions 1 , 2 ., Therefore , determining whether specified types or categories of sites , such as polymorphic 3 or substituted sites 4 within DNA or protein sequences , are concentrated in specific regions within DNA or protein sequences has become a key component of these analyses 5–8 ., For instance , detecting regions that feature heterogeneity in substitutions may provide valuable information on the structure and function of DNAs or proteins 9–13 ., Several parametric and nonparametric methods have been proposed and historically applied to sequence data ., Parametric methods include applications of a Fishers exact test to tallies of site types between regions , or of a likelihood ratio test to identify heterogeneous regions 14 , 15 ., Alternatively , several heuristic methods may be applied for this clustering 16 ., For example , UPGMA ( Unweighted Pair Grouping Method with Arithmetic-mean ) or NN ( Nearest Neighbor ) , are hierarchical methods that at each step combine the nearest 2 clusters into one new cluster ., Iteration of this step is continued until the number of clusters is one ., One of NNs variants , K-NN ( K-Nearest Neighbor ) , differs in its termination condition , stopping the iteration until the K clusters are identified , where K needs to be defined in advance ., Another heuristic approach , K-means , uses a partitioning algorithm to break data into K clusters , and also requires the number of clusters K as a prior knowledge ., When regions of a sequence that are expected to have heterogeneous frequencies of a site type may be specified in advance or the number of clusters to be identified is known a priori , these methods have high power to detect clustering 17 ., However , they require a priori assignment of partitions ., When no a priori expectation of cluster size or cluster number may be specified , extant studies have usually relied on “sliding window” methods 18–23 ., For example , Pesole et al . ( 1992 ) labeled invariable site as ‘1’ and variable site as ‘0’ , and applied a sliding window to identify whether ‘1’s are significantly clustered 24 ., Pesole et al . calculated a heuristic score based on the presence or absence of site types within a window that processes serially across the sequence of interest ., Advantages of sliding window methods include their intuitive conceptual basis and their striking output: an autocorrelated plot of the score that may be superimposed upon the sequence , providing a visual appraisal of the level of clustering at every site ., However , sliding window methods have two related major disadvantages 25 ., First , they generally offer only crude non-parametric means for statistical significance testing ., The autocorrelation of serial scores severely complicates attempts to develop more insightful parametric approaches to sliding window significance testing , making parameter estimation with confidence intervals either challenging or impossible ., Second , the need to specify a window size presents a user with a procedural ambiguity ., Without a unified statistical framework , there is no strong justification for selection of one window size over another ., In such a situation , it may even be tempting to invert the procedure of statistical inference and select a window size that produces an autocorrelated score plot consistent with a particular scientific hypothesis , as opposed to the valid procedure of selecting a window size by an objective statistical optimality criterion ., Because of these disadvantages of the sliding window methods , several nonparametric statistical methods that do not assume prior knowledge have been suggested or implemented to detect clustering in discrete linear sequences ., These methods include runs tests 26–28 and empirical cumulative distribution function ( ECDF ) statistics 29 , 30 ., Runs tests use the “longest unbroken run” between sites of interest as a test statistic for clustering , where a run is defined as consecutive length between events 26 ., This test statistic provides very weak power , because it uses very little of the relevant information about the phenomenon of interest , ignoring all runs other than the longest ., Statistics based on the longest two runs , longest three runs , or even on a summary of the full distribution of run lengths have been discussed , but remain weak tests ., For instance , the variance in distance between site types of interest may be calculated and used as a test statistic for the detection of clusters of sites , where a high variance is indicative of clustering 29 ., This test statistic incorporates information about the length of all the runs , but does not capture all of the relevant information: it discards all information about the relative position of runs of different lengths ., A sequence with all of its shorter runs in one region would be more clustered than one with short runs distributed evenly ., Currently , the most powerful nonparametric method is the ECDF ., It features the cumulative difference between the observed and expected proportion of variant sites to identify regions that differ from other regions in number of substitutions ., Under a null model that assumes no heterogeneous region ( s ) within sequences , this difference remains close to zero ., Its significant departure from zero is an indicator for rejecting the null model 29 , 30 ., Although ECDF has been used to detect heterogeneity in several studies 31–35 , its power can be affected by the location of the heterogeneous region 30 ., Moreover , a parametric method may perform even better across a wide range of datasets ., Most extant methods that have been proposed to detect heterogeneous clusters among sequences suffer from poor power to detect clustering when it is present ., The problem is made especially challenging by a tradeoff wherein increasing power to detect clustering also increases overparameterization or false positive rates ., Methods that have high power are prone to identify clustering even in random sequences , because even in short sequences , there are so many potential patterns of clustering to evaluate ., In this paper , we propose a hierarchical clustering method , model averaged clustering by maximum likelihood ( MACML ) , requiring no priori knowledge of cluster size or cluster count , that provides greater statistical power in detecting heterogeneous regions ., MACML adopts a divide-and-conquer approach to hierarchically detect heterogeneous regions and repeat similar analysis for each identified region , unlike most hierarchical methods that do not revisit clusters once they are constructed 17 , 36 , 37 ., To address issues of overparameterization , MACML employs model selection and model averaging techniques that lead to intuitively appealing profiles of sequence heterogeneity and that facilitate description of clustered sites in discrete linear sequences ., We describe MACML in detail and provide comparative results in the form of an in-depth evaluation of simulated datasets and an empirical sequence data set on polymorphisms in the Drosophila alcohol dehydrogenase gene ., To apply MACML to locate regional clusters with different specified site types requires a general input sequence X with N sites , denoted as ( 1 ) For example , to examine heterogeneity of substitution , an aligned set of homologous sequences is converted into X , in which each site is scored entries xi of 0 representing identity , and 1 representing a variant or variable site 30 ., Similarly , a sequence to be analyzed for detection of GC heterogeneity can be converted by setting G/C\u200a=\u200a1 and A/T, =\u200a0 . Notations used to describe our algorithm are summarized in Table, 1 . To test the performance of MACML and compare it to the most powerful extant method , ECDF , we simulated sequences for analysis for which the rates of variant sites were known a priori ., For each simulated sequence , we randomly generated the start and end positions of the cluster , positions of variant sites within the cluster region , and positions of variant sites within the non-cluster region ( see Figure 1 ) ., To avoid stochastic errors , we repeated simulations M\u200a=\u200a10000 times for each parameter combination ., Thus , each performance measure was determined from M replicates ., We retrieved the Drosophila alcohol dehydrogenase ( Adh ) gene within five species of Drosophila melanogaster species subgroup ( D . melanogaster , D . sechellia , D . simulans , D . yakuba and D . erecta ) from FlyBase 47 ., The aligned sequences of Drosophila Adh gene can be available at http://www . yale . edu/townsend/datasets . html ., The powers of MACML and ECDF were plotted against the percentage of variant sites within the cluster ( q ) under different numbers of variant sites ( n ) in Figure 3 and the corresponding accuracy and precision were plotted in Figure 4 ., Evaluating the methods based on their power to detect clusters within sequences with different q and n , MACML outperformed ECDF for nearly all the parameter combinations tested ( Figure 3 ) ., When n was very small , both methods exhibited extremely low power for detecting hot spots ( n\u200a=\u200a10 in Figure 3A ) ., At intermediate values of n , MACML and ECDF exhibited increasing power with q ( Figure 3B and 2C ) ., While ECDF approached the power of MACML when q was large , MACML remained more powerful across the full range of q ( Figure 3B to 2D ) ., The power of MACML and ECDF to detect cold spots was also low when n was small ( n\u200a=\u200a10 in Figure 3E ) ., When n increased to 50 , the power of MACML and ECDF peaked at intermediate values of q ( Figure 3F ) ., At higher levels of n\u200a=\u200a100 ( Figure 3G ) and n\u200a=\u200a200 ( Figure 3H ) , ECDF continued to peak at intermediate values of q , whereas the power of MACML continued to rise with q ., Across the parameter ranges examined , MACML consistently exhibited greater power than ECDF ., The accuracy and precision of MACML and ECDF were estimated by the Kullback-Leibler ( KL ) divergence , which is a measure of the difference between the expected and estimated distributions of variant rates ., In assessing the accuracy based on the KL divergence , therefore , there are three potential scenarios: a good match between the estimated and expected variant rates when a KL divergence is near zero , an underestimation of variant rates when KL divergence is positive , and an overestimation of variant rates when KL divergence is negative ., The precision based on the KL divergence is also better when it is closer to zero ., Unlike the accuracy , precision based on the KL divergence cannot be negative ( Equation 12 ) ., Evaluating the accuracy and precision based on the KL divergence , MACML performed better than ECDF for most of the cases examined ( Figure 4 ) ., The accuracy and precision of MACML and ECDF for detecting hot spots were very good ( near zero ) when n was small ( Figure 4A ) ., When n became large , MACML exhibited good accuracy and precision , whereas the accuracy and precision of ECDF diverged positively from zero with increasing q ( Figure 4B to 3D ) ., This divergence was augmented when n was extremely large ( Figure 4D ) ., When n is small ( n\u200a=\u200a10 in Figure 4E ) , both MACML and ECDF also exhibited good accuracy and precision for the detection of cold spots ., At large values of n ( Figure 4F to 3H ) , ECDF exhibited good accuracy and precision only when q was smaller ( 10% ) or larger ( 90% ) ., At intermediate values of q , the accuracy of ECDF diverged from the ideal negatively ., The precision of ECDF diverged from the ideal as well ., This divergence was augmented when n was extremely large ( n\u200a=\u200a200 in Figure 4H ) ., In summary , MACML exhibited good accuracy and precision for nearly all tested cases ., The powers of MACML and ECDF were plotted against the ratio of variant rates within cluster to outside of cluster in Figure 5 , and the corresponding accuracy and precision were plotted in Figure 6 ., The difference in power between MACML and ECDF was least remarkable for the detection of cold spots ( Figure 5A ) ., At values of the ratio of variant rates within cluster to outside of cluster ranging from 0 . 3 to 0 . 9 , differences in power between both methods were relatively small , whereas at values of the ratio <0 . 3 , MACML showed much greater power to detect cold spots than did ECDF ( Figure 5A ) ., The power of MACML to detect hot spots consistently increased with increasing ratio ( Figure 5B ) ., Although the power of ECDF increased with the ratio as well , its power was much lower than the power of MACML across the examined ranges of values of the ratio ( Figure 5B ) ., MACML provided good accuracy and precision ( near zero ) for detecting cold spots , whereas the accuracy of ECDF diverged negatively and the precision of ECDF diverged from the ideal as well ( Figure 6A ) ., This divergence was more notable at values of the ratio <0 . 7 ( Figure 6A ) ., With regard to hot spots , the accuracy and precision of ECDF diverged positively across values of the ratio from 2 to 10 ( Figure 6B ) ., As the ratio was increased , this divergence became more remarkable ., In contrast , MACML exhibited better accuracy and precision for most of the examined cases ( Figure 6B ) ., According to their definitions , the ratio of variant rates within cluster to outside of cluster\u200a=\u200a1∶1 , q\u200a=\u200a0% , or q\u200a=\u200a100% represent sequences with entirely randomly located substitutions under the null model ., Therefore , we compared three criteria adopted by MACML and examined their errors of overparameterizing the clustering model when no clustering was imposed during the sequence generation ., MACML and ECDF demonstrated high overparameterization and false positive rates , respectively ( Table 2 ) ., The overparameterization rate of MACML markedly exceeded the false positive rate of ECDF for n\u200a=\u200a10 , n\u200a=\u200a100 and n\u200a=\u200a200 ., Implementing the AIC and AICc did little to moderate overparameterization , whereas implementing BIC significantly moderated overparameterization ., Implementing the BIC did not bring overparameterization down to the false positive rate of ECDF for n\u200a=\u200a10 , 100 , and 200 , but did limit the overparameterization rate to approximately the false positive rate of ECDF for sequences with n\u200a=\u200a50 ., The powers of MACML and ECDF were plotted against sequence length in Figure 7 and the corresponding accuracy and precision were plotted in Figure, 8 . When sequence length increased from 100 to 1000 sites , MACML and ECDF provided decreasing power to detect both hot spots ( Figure 7A ) and cold spots ( Figure 7B ) ., This decrease was more prominent for MACML than for ECDF ., Nonetheless , MACML outperformed ECDF for most of these cases ., The accuracy and precision of MACML and ECDF varied little across all values of sequence length ., With increasing sequence length , the accuracy of ECDF diverged from zero positively for hot spots and diverged slightly negatively for cold spots ., The precision of ECDF diverged from the ideal positively for both hot spots and cold spots ( Figure 8A and 7B ) ., Overall , MACML exhibited better accuracy and precision than ECDF as sequence length increased from 100 to 1000 ( Figure 8 ) ., The powers of MACML and ECDF were plotted against the number of clusters in Figure, 9 . Under the parameters examined for multiple clusters ( see Materials and Methods ) , MACML and ECDF performed similarly when the sequence had only one cluster to be detected ., However , when the number of clusters ranged from 2 to 10 , ECDF was unable to detect more than one cluster , whereas MACML had significant power to detect multiple clusters , especially for large values of n ., In general , the power of MACML was limited for small values of n\u200a=\u200a10 ( Figure 9A ) and n\u200a=\u200a50 ( Figure 9B ) , but much greater for large values of n\u200a=\u200a100 ( Figure 9C ) and n\u200a=\u200a200 ( Figure 9D ) ., We applied MACML to detect heterogeneous clusters of polymorphisms within the Drosophila Adh gene and to profile potential for polymorphism for each site based on model selection and model averaging , respectively ., Identified clusters as well as profiles of the potential for polymorphism were plotted against sequence coordinate ( Figure 10 ) ., As expected , profiles of potential for polymorphism based on model selection ( Figure 10A and 9C ) are highly discrete , whereas smoother , continuous profiles are produced based on model averaging ( Figure 10B and 9D ) ., When using BIC , MACML detected two clusters along the Adh gene and both are cold spots residing between sites 98 and 189 and between sites 26 and 70 ( Figure 10A and 9B ) ., In addition to these two cold spots , when using AIC or AICc , MACML also identified two hot spots between sites 80 and 84 and between sites 212 and 218 ( Figure 10C and 9D ) ., In contrast , ECDF detected only one cold spot between sites 98 and 211 ( data not shown ) , consistent with previous applications of the method 29 , 30 ., Detailed clustering results for the Adh gene are summarized in Table 3 ., For the AIC or AICc , the four detected clusters all deviate significantly from the null model ( ΔAIC<0 and ΔAICc<0 in Table 3 ) ., When sample size is large , like sequence from sites 0 to 253 , the ΔAICc asymptotically approaches ΔAIC , and thus their values are nearly same ., However , for a smaller sample size , for example , when detecting sub-sequence from sites 71 to 97 , ΔAICc is much larger than ΔAIC ., By contrast , BIC incorporates a heavier penalty than AIC or AICc and ΔBIC>0 indicated no significant cluster among sub-sequences from sites 71 to 97 or from 190 to 253 , whereas AIC and AICc identified two clusters along these two sub-sequences ., The power to detect heterogeneous clustered sites within sequences depended in moderately complex ways on the parameters we examined in this report ., Consistent with expectations , our results show that the power of MACML to detect hot spots and cold spots increased with increasing percentage of variant sites within the cluster ( Figure 3 ) ., Across simulations comparing different percentages of variant sites within the cluster , MACML exhibited both high accuracy and high precision: the estimated variant rates within and outside clusters were close to the expected ones across all parameter combinations ( Figure 4 ) ., In contrast to MACML , ECDF performed more variably across different percentages of variant sites within the cluster ., This inconsistency of performance agrees well with our theoretical analysis on ECDF ( Text S1 ) as well as with results from a previous study 30 ., The hot spots and cold spots estimated by ECDF tend to be narrower than the simulated hot spots and cold spots 30 ., The misattributed region between the boundary of the estimated hot or cold spot and the corresponding boundary of the simulated hot or cold spot generally gives rise to much greater KL divergence than any other region of the sequence ., Thus , the KL divergence of the full sequence tends to be dominated in direction and magnitude by the KL divergence of the region between these boundaries , a region that is usually present as a consequence of the bias in estimation of the width of hot and cold spots ., Accordingly , positive divergence from perfect accuracy and precision for hot spots ( Figure 4A to 3D ) follows from underestimation of the variant rate of this region ., Likewise , negative divergence from perfect accuracy and positive divergence from perfect precision for cold spot ( Figure 4E to 3H ) follows from overestimation of the variant rate of this region ., Across a range of ratios of variant rates within the cluster to outside of the cluster , MACML and ECDF exhibit similar trends in power , but different trends in accuracy and precision ., With both methods , a significant difference between variant rates within the cluster and outside of the cluster leads to greater power , and nearly equal rates for all sites results in lower power ( Figure 5 ) ., The KL divergence measure of the accuracy of ECDF is negative for cold spots and positive for hot spots , respectively ( Figure 6 ) ., When the variant rate inside of the cluster approaches the variant rate outside of the cluster , estimated and actual variant rates are very close for any cluster model ., Therefore , the accuracy and precision of ECDF approach those of MACML , consistent with simulation results ( Figure 6 ) ., In contrast , as variant rates within the cluster diverge from rates outside the cluster , MACML produces incrementally better accuracy and precision across all parameter combinations ( Figure 6 ) ., Both MACML and ECDF exhibit decreasing power with increasing sequence length ( Figure 7 ) , presumably as a consequence of the decreasing proportion of variant sites relative to sequence length ., Increasing sequence length with a fixed number of variant sites is equivalent to decreasing the number of variant sites with a fixed sequence length ., Therefore , it is consistent that the power decreases with decreasing variant sites in Figure 3 ., This relationship between variant sites and power also agrees well with the results observed when varying the number of clusters ( Figure 9 ) , with the additional note that ECDF fails to detect more than one cluster ., It is notable that simulations performed by Tang and Lewontin 30 were less general in scope than ours ., That is , in Tang and Lewontin 30 , the heterogeneous cluster was always centered and the two regions flanking the cluster were always equal in length ., As noted by Tang and Lewontin , the power of ECDF is affected when the cluster moves off center 30 ., In our simulations , the starting position and ending position of cluster are randomly generated , leading to a random location of the cluster and thus to an unequal length of the two flanking regions ( see details in Materials and Methods ) ., For these reasons , our simulations that incorporated random positions of clusters yielded different results in terms of success detecting multiple clusters than were yielded by the simulations of Tang and Lewontin 30 ., False positive rates and overparameterization for clustering models were high , as expected as a consequence of the large number of potential cluster boundary sets that are possible ., Powerful methods for this class of problem are expected to display high false positive rates , a tradeoff that is natural in statistical inference ., Although ECDF presents lower false positive rates , MACML achieves more power than ECDF to reject the null hypothesis when it is not true ( Figures 3 , 4 and 6 ) ., Moreover , MACML achieves markedly greater accuracy and precision of variant rates as determined by the KL divergence ( Figures 3 , 5 and 7 ) , demonstrating the marked superiority of MACML in selecting the best model of variant rates across a discrete linear sequence ., Furthermore , MACML is more capable of detecting multiple clusters among sequences , as demonstrated by simulation ( Figure 9 ) and by application to the empirical data ( Figure 10 ) ., Unlike ECDF , which is not integrated into a model selection framework , MACML adopts AIC , AICc and BIC for model selection ., To clarify the differences observed implementing these diverse criteria , the different penalties for additional parameterization that they entail may be compared ., Based on the clustering model , two parameters ( cs and ce ) are evaluated ( from which p0 and pc can be calculated ) ., Therefore , the number of parameters under the clustering model is two , whereas the number under the null model is zero ., From equations 4–6 , then , ( 11 ) ( 12 ) ( 13 ) where l is sample size , that is , ( sub- ) sequence length ., The values of lnLc–lnL0 may be plotted against sample size ( Equations 11–13 , Figure 11 ) ., AIC yields constant penalties for all values of sample size ., For smaller sample size , AICc yields larger penalties than AIC or BIC ., When sample size increases to large numbers , the penalty of AICc approaches AIC , and BIC produces much larger penalties than AICc ., For a given value of lnLc–lnL0 , the three criteria are most likely to give different results with regard to rejection of the null model ., The three lines plotted corresponding to the three different criteria in Figure 11 may be helpfully related to the results of our application of MACML to the Adh gene ., MACML started by detecting a cluster from site 0 to 253 ., The sample size was 254 , and the corresponding value of lnLc–lnL0 was 6 . 53 ( Table 3 ) ., This cluster is represented by a point ( 254 , 6 . 53 ) , located above all three lines ., This location signifies that the three criteria all reject the null model ., After locating the first cluster , MACML proceeded to detect clusters along sub-sequences from 0 to 97 , from 98 to 189 , and from 190 to 253 , until all possible sub-sequences had been tested ., As a consequence , it identified several clusters ., Two of them are located above the three lines , signifying that all three criteria reject the null model ., The remaining two points are located below the BIC line and above the other lines , signifying that BIC does not reject the null model , but that the rest do ( Figure 11 ) ., This graphical analysis clarifies results in which BIC identified only two cold spots , whereas the other criteria identified an additional two hot spots ( Figure 11 and Table 3 ) ., The Drosophila Adh is the most studied enzyme that catalyzes the oxidation of alcohols to aldehydes/ketones 48 ., It has been extensive reported that several functionally important residues reside in the Adh gene: tyrosine-152 , lysine-156 and serine-139 are conserved in homologous dehydrogenases and have important roles in catalysis 49–53; glycine-130 , glycine-133 and glycine-184 contribute substantially to the structure of the active form 50; and aspartic acid-64 lies within a coenzyme-binding domain 51 ., As shown in Figure 10 and Table 3 , these residues were all clustered into the cold spots by MACML , indicating not only their functional conservation and relevance , but also the extent of the region of near-neighbor amino acids that are also conserved ., Near-neighbors may be conserved due to their structural and biochemical effects on the known function of these residues ., In addition , according to its gene structure , two introns in the Adh gene reside between the nucleotide sequences coding for residues 32 and 33 and between the nucleotide sequences coding for residues 167 and 168 54 , 55 ., Therefore , the two cold spots identified by MACML extending from residues 26 to 70 and from residues 98 to 189 indicate conservation around the introns ., Heterogeneity of variant rates among specified site types is thought to commonly occur 56–59 and may derive from many sources , including functional constraint , gene structure , 3D protein structure , composition bias , mutation bias or recombination 1 , 18 , 34 , 60–62 ., As indicated by our results based on the simulated data and real data , MACML , equipped with model selection and model averaging , features smooth and continuous profiles of variant rates for each site , and is more accurate and more informative for the detection of multiple clusters among sequences ., Therefore , MACML furnishes broad utility for any computational analyses of heterogeneous discrete linear sequences and provides valuable information to aid for a better understanding of the structure and function of DNAs or proteins ., In addition , MACML can be applied to a broad range of applications ., For example , MACML would be appropriate for determining whether components of any multicomponent polymer have a clustered structure 33 , 63 ., It can also be used to detect compositional heterogeneity within sequences 64–66 ( e . g . , heterogeneous GC content by setting G/C\u200a=\u200a1 and A/T\u200a=\u200a0 ) ., Moreover , MACML may provide a framework upon which future modeling of the substitution process may be overlain , assessing heterogeneity in selective pressure acting on different coding sequence regions 60 , 67–70 and detecting fast-evolving regions in noncoding sequences 71 , 72 ., Here we have presented a method , MACML , to detect clustering of a site type in discrete linear sequences ., MACML features maximum likelihood estimation , model selection criteria ( AIC , AICc , and BIC ) and model averaging to profile sequence heterogeneity ., It employs a divide-and-conquer approach to hierarchically detect multiple clusters within sequences , without requiring a priori knowledge for cluster size or number ., We compared MACML with the most powerful competing method , the ECDF , by exploring a full range of parameter space using computer simulations , and by performing an analysis of empirical data ., Our comparative results show that across a wide range of parameter combinations , MACML outperforms ECDF not only by exhibiting greater power to detecting hot spots and cold spots ., Thus , it represents a powerful exploratory tool for profiling clustering in discrete linear sequences ., Although discoveries using MACML should be considered tentative , it yields greater resolution than any other method , providing a significant advance for the analysis of clustering of sites within discrete linear sequences .

Introduction, Materials and Methods, Results, Discussion

A major analytical challenge in computational biology is the detection and description of clusters of specified site types , such as polymorphic or substituted sites within DNA or protein sequences ., Progress has been stymied by a lack of suitable methods to detect clusters and to estimate the extent of clustering in discrete linear sequences , particularly when there is no a priori specification of cluster size or cluster count ., Here we derive and demonstrate a maximum likelihood method of hierarchical clustering ., Our method incorporates a tripartite divide-and-conquer strategy that models sequence heterogeneity , delineates clusters , and yields a profile of the level of clustering associated with each site ., The clustering model may be evaluated via model selection using the Akaike Information Criterion , the corrected Akaike Information Criterion , and the Bayesian Information Criterion ., Furthermore , model averaging using weighted model likelihoods may be applied to incorporate model uncertainty into the profile of heterogeneity across sites ., We evaluated our method by examining its performance on a number of simulated datasets as well as on empirical polymorphism data from diverse natural alleles of the Drosophila alcohol dehydrogenase gene ., Our method yielded greater power for the detection of clustered sites across a breadth of parameter ranges , and achieved better accuracy and precision of estimation of clusters , than did the existing empirical cumulative distribution function statistics .

The invention and application of high-throughput technologies for DNA sequencing have resulted in an increasing abundance of biological sequence data ., DNA or protein sequence data are naturally arranged as discrete linear sequences , and one of the fundamental challenges of analysis of sequence data is the description of how those sequences are arranged ., Individual sites may be very sequentially heterogeneous or highly clustered into more homogeneous regions ., However , progress in addressing this challenge has been hampered by a lack of suitable methods to accurately identify clustering of similar sites when there is no a priori specification of anticipated cluster size or count ., Here , we present an algorithm that addresses this challenge , demonstrate its effectiveness with simulated data , and apply it to an example of genetic polymorphism data ., Our algorithm requires no a priori knowledge and exhibits greater power than any other unsupervised algorithms ., Furthermore , we apply model averaging methodology to overcome the natural and extensive uncertainty in cluster borders , facilitating estimation of a realistic profile of sequence heterogeneity and clustering ., These profiles are of broad utility for computational analyses or visualizations of heterogeneity in discrete linear sequences , an enterprise of rapidly increasing importance given the diminishing costs of nucleic acid sequencing .

computational biology/population genetics, computational biology/macromolecular structure analysis, computer science/applications, genetics and genomics/comparative genomics, computational biology/sequence motif analysis, evolutionary biology/evolutionary and comparative genetics, computational biology/comparative sequence analysis, computational biology/molecular genetics, molecular biology/molecular evolution, molecular biology/bioinformatics, computational biology/macromolecular sequence analysis, computational biology/genomics, computational biology, mathematics/statistics, genetics and genomics/bioinformatics

journal.pgen.1003841

Whole Genome Sequencing Identifies a Deletion in Protein Phosphatase 2A That Affects Its Stability and Localization in Chlamydomonas reinhardtii

Forward genetics allows the identification of mutants with phenotypes of interest and the mechanistic understanding of biological processes 1 , 2 ., While gene lesions generated by insertional mutagenesis can be identified by Southern blot analysis 3–6 or PCR-based approaches 7–9 , identification of mutations induced by radiation or chemical mutagenesis rely on time-consuming meiotic mapping 10–14 ., Recently , single nucleotide polymorphism ( SNP ) discovery by whole genome sequencing ( WGS ) provides a faster and more efficient method to identify causative mutations 15–18 ., However , in model organisms such as Arabidopsis thaliana , the number of SNPs can vary from 2 , 000 to 900 , 000 , depending on the strain background 19 ., In Caenorhabditis elegans , the number of SNPs between two strains is ∼100 , 000 16 ., In the unicellular biflagellate green alga Chlamydomonas reinhardtii , ∼38 , 000 SNPs were identified in individual mutant strains 18 ., Therefore , identification of the causative SNP from a large number of SNPs remains a challenge ., In Chlamydomonas , a model organism for the study of flagellar function , photosynthesis , biofuels , and sex determination , the causative genes in several hundred mutant strains generated by radiation or chemical mutagenesis remain unidentified 20 ., UV mutagenesis mutant screens generated 12 impotent ( imp ) mutant strains that have either abolished or reduced mating efficiency 21 , 22 ., In Chlamydomonas , the sex of a cell is controlled by two alleles , plus or minus , at the mating-type ( MT ) locus 23 ., The differentiation from exponentially growing vegetative cells to gametes is triggered by nitrogen starvation via an unknown mechanism ., When gametes of the opposite mating-types are mixed together , they agglutinate via flagellar membrane-associated proteins , agglutinins , to trigger a mating signal transduction pathway ., This signal cascade leads to cell fusion and the formation of zygotes ., Among the previously characterized imp mutant strains , imp2 , imp5 , imp6 , imp7 , and imp9 are allelic and encode SAG1 , which is the plus agglutinin 24 ., The imp10 and imp12 mutant strains encode SAD1 , the minus agglutinin 24 ., The imp8 strain is defective in O-glycosylation and is allelic with the GAG1 locus 25 ., The imp1 and imp11 mutant strains map within the MT locus 26–29 and carry mutations in FUS1 in plus and MID1 in minus cells , respectively ., Only imp3 and imp4 remain unidentified among the original collection of impotent mutants ., Unlike the other imp strains that abolish mating ( <1% ) , the mating efficiency of imp3 and imp4 strains varies from 10% to 50% , in contrast to >80% in wild-type cells 1 hour after mixing of the gametes ., Neither mutation is linked to the MT locus or to the other 21 ., Saito et al . 30 suggested that activation of uncharacterized flagellar adenylate cyclases is blocked in imp3 cells , and IMP3 is required in the mating signaling pathway ., However , the causative genes in imp3 and imp4 remain unidentified due to their partial , weak mating phenotype and the difficulty to map this phenotype ., Serine/threonine phosphorylation is generated by 300–400 kinases but is reversed by a relative small number of phosphatases ., The serine/threonine phosphatase , protein phosphatase 2A ( PP2A ) , plays an important role in signaling pathways ., It is a ubiquitous enzyme that is involved in diverse cellular processes; they include cell cycle control , cell growth , microtubule stability , and signaling 31 ., The PP2A heterotrimeric holoenzyme contains 3 subunits; they are a catalytic subunit ( C subunit ) , a scaffold subunit ( A subunit ) , and a regulatory subunit ( B subunit ) ., The catalytic subunit ( PP2Ac ) is highly conserved across species and it shares significant sequence similarity to the PP4 and PP6 protein phosphatases 32 ., The catalytic activity of PP2Ac can be modulated by post-translational modifications that include phosphorylation/dephosphorylation in the conserved C-terminal motif T304PDY307FL309 on T304 and Y307 and methylation/demethylation of L309 33 ., The scaffold subunit of PP2A contains multiple HEAT repeats that confer conformational flexibility to both the catalytic subunit and the regulatory subunit 32 ., The regulatory subunit of PP2A falls into four distinct families , which are known as B ( PR55 ) , B′ ( B56 or PR61 ) , B″ ( PR72 ) , and B′″ ( PR93/PR110 ) ., It is believed that different families of the B subunit target PP2A to different cellular locations and bind to different substrates 32 , 33 ., In Chlamydomonas , sequence similarity reveals four catalytic subunits , two scaffold subunits , and five regulatory subunits 34 ., In this study , we took advantage of whole genome sequencing of 16 different wild-type and mutant strains to generate a SNP/indel library ., SNP/indel comparison , in conjunction with meiotic mapping , allowed the quick identification of the causative mutation in the imp3 mutant strain , which contains a C-terminal three amino acid deletion in the conserved TPDYFL motif of a PP2A catalytic subunit ( PP2A3 ) ., The deletion of YFL affects not only the stability of PP2A3 , but also the accumulation of PP2A3 around the basal body area ., A previous study using 101-bp paired-end Illumina sequencing of an IFT80 mutant strain , NG30/ift80 , revealed that over 38 , 000 SNPs/indels are present compared to the Chlamydomonas reference genome 18 ., It was a challenge to identify the causative mutation from such a large number of changes ., We reasoned that if changes are found in other unlinked mutant strains or in wild-type strains , they are not causal and could be eliminated from further analysis ., Therefore , a collection of changes from multiple strains would be necessary to remove as many non-causative changes as possible to reveal the causative mutation in a given mutant strain ., To build a library of changes , we first sequenced four wild-type strains ( CC-124 , CC-125 , isoloP ( CC-4402 ) , and isoloM ( CC-4403 ) ) ., In Chlamydomonas , the major laboratory strains CC-124 and CC-125 were first isolated from a single diploid zygote , 137c , in 1945 35 and these strains have been used as the parents in many mutant isolations ., CC-125 is also the background strain of CC-503 , the strain used for the Chlamydomonas reference genome assembly ., CC-124 carries a minus mating-type locus ( mt− ) and CC-125 is mating-type plus ( mt+ ) ., The other two wild-type strains , isoloP ( mt+ ) and isoloM ( mt− ) , were generated by crossing CC-124 by CC-125 to obtain meiotic progeny ., Several progeny from this cross that gave the fastest and highest percentage of mating were backcrossed to CC-124 ., This procedure was repeated ten times in an attempt to obtain isogenic strains that were named isoloP and isoloM ., Sequencing of these wild-type strains identifies 13 , 000 to over 100 , 000 changes relative to the reference genome ., We also sequenced is a highly polymorphic strain S1C5 ( CC-1952 ) that is frequently used in meiotic mapping with molecular markers ., Over 2 million changes are identified from this strain ., In addition , since we were interested in identifying the causative mutations from a variety of mutant strains , we also performed whole genome sequencing on ten mutant strains that were generated by either chemical or UV-mutagenesis ., Five of them ( fla18 , fla24 , fla9 , uni1 , ift80 ) have flagellar assembly defects , four ( ida3 , pf23 , pf7 , pf8 ) have motility defects , and one has a mating defect ( imp3 ) ., One additional mutant strain , cnk10 , which has a flagellar assembly defect , was generated by insertional mutagenesis of the CC-125 strain ( Lin and Dutcher , unpublished ) ., The number of changes in individual mutant strains varies from 22 , 000 to over 150 , 000 ( Table 1 ) ., The sequencing coverage of individual strains ranges from 26× to 162× ( Table 1 ) ., Overall , 2 , 557 , 197 changes are included in this SNP/indel library and it is available at http://stormo . wustl . edu/SNPlibrary/ ., After collecting the SNPs/indels , we analyzed the distribution of the changes across the 17 chromosomes relative to the reference genome 36 ( Figure 1 ) ., In the wild-type strain CC-125 ( mt+ ) , changes are spread evenly across all chromosomes from the reference genome ( Figure 1A , Table 2 ) ., The wild-type strain CC-124 ( mt− ) , which came from the same zygote as CC-125 , contains 100 , 737 changes , which is about eight times the number of changes ( 13 , 218 ) found in CC-125 ., Around 90% of changes found in CC-124 are concentrated on five chromosomes: 3 , 6 , 12 , 16 , and 17 ( Figure 1A , Table 2 ) ., A detailed analysis of numbers of SNPs/indels every 100 kb along the chromosomes in CC-124 reveals that the polymorphisms are not distributed evenly across these five chromosomes ( Figure 1B ) ., On chromosome 3 , most SNPs/indels are between 8 . 5 Mb and 9 . 1 Mb ., On chromosome 6 , most SNPs/indels are within the first 1 . 9 Mb , which contains the MT locus ., The mating-type locus is known to be polymorphic between the two sexes 37 ., CC-124 carries the MT minus locus , which is not shared with the reference strain CC-503 ., On chromosome 12 , most changes are observed between 9 . 0 Mb to 9 . 8 Mb ., On chromosome 16 , three distinct regions of SNPS/indels lie between 0 . 9–1 . 0 Mb , 1 . 5–2 . 0 Mb , and 6 . 4–7 . 8 Mb ., A large number of changes within the 0 . 9–1 . 0 Mb were observed in a previous study of Chlamydomonas strains , ift80 and ac17 18 ., On chromosome 17 , most changes are observed between 0 . 3 Mb to 1 . 5 Mb ., The other two wild-type strains , isoloP ( mt+ ) and isoloM ( mt− ) , which are meiotic progeny of CC-124 and CC-125 after ten rounds of backcrosses to CC-124 , were expected to show difference only on chromosome 6 , which contains the MT locus ., However , comparison of the sequence of these two strains indicates that they are not isogenic on chromosomes 3 and 17 ., The isoloM strain maintains large numbers of changes from the CC-124 parent on chromosome 3; isoloP contains large numbers of changes inherited from the CC-124 parent on chromosome 17 ( Figure 1A , Table 2 ) ., We performed the same analysis on the ten mutant strains , with the exclusion of fla18 , because the sequenced strain came from a cross between the fla18 mutant strain and S1C5 ., Ninety-six percent of fla18 SNPs/indels are found in the S1C5 strain ., Prior to Illumina sequencing , the ten strains were crossed to either CC-124 or CC-125 at least once ., An accumulation of changes on chromosomes 3 , 6 , 12 , 16 , and 17 are observed in most strains ( Figure 1C ) ., To ask whether these changes are the same as found in CC-124 , we subtracted SNPs/indels found in CC-124 from individual strains ., The numbers of changes drop dramatically from 10 , 000∼35 , 000 to 1 , 000∼5 , 000 in almost all strains ( Figure 1D ) ., This suggests that changes in these strains are likely come from the CC-124 parent ., In comparison , removal of CC-125 changes from these strains does not have an obvious effect on numbers of changes in these strains ( Figure 1E ) ., There is no correlation with the position of the causative mutant and an accumulation of SNPs along the chromosomes ., One mutant strain , uni1 , has a distinct distribution of changes along all chromosomes when compared to other strains ( Figure 1C ) ., Accumulation of changes is found on chromosomes 1 , 2 , 11 , and 16 ., Removal of CC-124 or CC-125 changes from uni1 has no significant effect on the distribution ( Figure 1D and 1E ) ., A comparison between 30 , 958 changes on chromosome 16 found in CC-124 and 56 , 192 changes on chromosome 16 found in uni1 indicates that only 8 , 229 changes are common between these two strains ., Thus , the SNPs/indels found in uni1 have significantly different distribution than all other strains we analyzed ., This mutant strain was generated from either strain 89 or 90 ( CC-1009 or CC-1010 ) 38 ., Pröschold et al . 39 categorized the common used laboratory Chlamydomonas strains into three basic sublines based on several criteria , including their ability to utilize nitrate , the mating-type locus , the number of rDNA repeats , and the presence of cell wall digestion metalloproteases ( autolysins ) ., Strains 89 and 90 belong to Subline II and CC-124 and CC-125 belong to Subline III ., They have difference in all criteria described above ., Thus it is not surprising to observe the difference of SNP/indel accumulations between uni1 and 137c strains ., The 162× sequencing coverage of imp3 reveal 91 , 066 changes in this mutant strain , which came from a 137c background ., Comparison and subtraction between imp3 and all other 15 strains finds 7 , 092 ( 7% ) of these changes are unique to imp3 ( Table 1 ) ., Out of 954 changes in predicted exons , 145 are predicted to be synonymous changes and were excluded from further analysis ., Changes vary from 1 to 297 on individual chromosomes ( Table 3 ) ., In order to identify the causative mutation , we meiotically mapped the imp3 mutant strain ., The imp3 phenotype confers reduced mating efficiency , a phenotype that is challenging to analyze quickly ., When wild-type gametes mate , the zygotes develop thick cell walls and form a dark green , multi-layered sheet of cells called the pellicle ( Figure 2A ) ., The mating between imp3 mt+ and imp3 mt− do not form the thick pellicle sheet observed in wild-type cells or in mating between imp3 and wild-type cells ( Figure 2A ) ., Observation under the dissecting microscope reveals dark thick multi-layer pellicle between imp3 and wild-type gametes ( Figure 2B ) ., Mating between two imp3 strains produces a single cell-layer pellicle that is light in color ( Figure 2C and 2D ) ., Individual progeny from 30 complete tetrads from a cross between imp3 and wild-type cells were tested for this phenotype by mating with imp3 mt+ and imp3 mt− tester strains ., All 30 tetrads showed 2∶2 segregation of the single layer pellicle phenotype ., This phenotype made it easy to distinguish between imp3 and IMP3 cells; this assay facilitated the molecular mapping of the IMP3 locus ., To map the imp3 mutation , meiotic progeny from imp3 crossed by the highly polymorphic strain S1C5 were obtained ., Progeny from twenty tetrads that showed the single layer pellicle phenotype were mapped with dCAPS markers ( Table S1 ) ., The single layer pellicle phenotype showed very tight linkage ( 18 parental: 0 recombinant ) to the SCA8-2 marker , which maps to ∼6 . 6 Mb on chromosome 2 in a previous version of the genome assembly ( v4 ) and maps to ∼3 . 5 Mb on chromosome 9 in the latest version ( v5 . 3 ) of the genome assembly ., To fine map the imp3 mutation , additional progeny from over 100 tetrads were used for mapping ., All three markers , 2-98 , SCA8-2 , and 55-193 , which are about 17 , 8 , and 3 cM away from the imp3 mutant , map to chromosome 2 in v4 genome assembly ( Table S1 ) ., Both 2-98 and 55-193 are at ∼6 . 6 Mb and ∼7 . 1 Mb on chromosome 2 in v5 . 3 genome assembly , respectively ( Table S1 ) ., Thus , we believe that genomic DNA with an unknown length around the SCA8-2 marker is misassembled in v5 . 3 genome assembly to chromosome 9 and the imp3 mutation maps to chromosome 2 ., However , since we were unclear whether the imp3 mutation is misassembled in v5 . 3 genome assembly , we examined polymorphisms in predicted exons on both chromosomes 2 and 9 ., Two 3 nucleotide-insertion changes at ∼2 . 3 Mb on chromosome 2 in v5 . 3 genome assembly are found in imp3 ( Table 3 ) ., Since both changes are over 4 Mb away from the 2-98 and 55-193 markers , they were eliminated from further study ., Two changes are found on chromosome 9 in v5 . 3 genome assembly ( Table 3 ) and they are both within the mapping region of ∼8 cM defined by the SCA8-2 marker ., Both changes map to chromosome 2 in the v4 genome assembly ., The first SNP change , which maps to position 4 , 049 , 338 on chromosome 9 , has 5 Illumina sequencing reads and a Phred quality score of 16 . 9 ., It is a G to A change that causes a non-synonymous change from R ( cGg ) to Q ( cAg ) in a RegA/RlsA-like protein ( g9750 ) ., The exact same change is observed in an aflagellate mutant cnc1 not included in the SNP/indel library ( Dutcher and Nauman , unpublished ) and thus is unlikely to be the causative change in the imp3 mutant strain ., The second change , which maps to position 3 , 721 , 280 on chromosome 9 , has 122 reads and a Phred quality score of 214 ., It is a deletion of 9 nucleotides immediately before the stop codon of a PP2A catalytic protein ( PP2A3 , g9684 ) and the deletion is predicted to remove the last 3 amino acids , YFL , which are conserved in almost all type 2A phosphatases ( PP2A , PP4 , and PP6; Figure 3B , S1 , and S2 ) ., This change was confirmed by Sanger sequencing ., A previous study using sequence similarity indicated that the Chlamydomonas genome contains four potential PP2A catalytic subunits , PP2A-1c ( g4366 ) , PP2A3 ( g9684 ) , PP2A-c4 ( Cre12 . g494900 ) , and PPA1 ( Cre06 . g308350 ) 34 ., Due to the sequence similarities observed among PP2A , PP4 , and PP6 in all organisms 32 , we asked whether the four Chlamydomonas proteins are PP2A catalytic subunits using phylogenetic analysis ., A phylogenetic tree was built based on 55 PP2A , PP4 , and PP6 protein sequences from green algae ( Chlamydomonas reinhardtii , Chlorella variabilis , Micromonas , Ostreococcus lucimarinus , Ostreococcus tauri , and Volvox carteri ) , yeast ( Saccharomyces cerevisiae ) , land plants ( Arabidopsis thaliana and Zea mays ) , invertebrates ( Caenorhabditis elegans and Drosophila melanogaster ) , and mammals ( Mus musculus and Homo sapiens ) ( Table S2 , Figure 3A and S1 ) ., This phylogenetic tree shows that Chlamydomonas PP2A-1c and PP2A3 are PP2A-like proteins ., PP2A-c4 is a PP4-like protein and PPA1 belongs to the PP6 family ., Within the PP2A family , four subgroups are distinguished using a bootstrap analysis ( Figure 3A ) ., PP2A proteins from invertebrates , mammals , and a green alga Chlorella form subgroup 1 in the PP2A family ., The subgroup 2 is composed of PP2A proteins from land plants and yeast ., PP2A proteins from land plants and green algae are found in subgroup 3 ., The subgroup 4 , which includes PP2A3 , is a green algal-specific subgroup ., Protein sequences of all 26 proteins from the PP2A family were aligned ., The alignment reveals that while subgroups 1 , 2 , and 3 have the conserved T304PDYFL309 C-terminus , the proteins in the fourth subgroup do not have the conserved T304 , instead , it is replaced with V , C , or M ( Figure 3B and S2 ) ., This suggests that T304 is not conserved in the green algal-specific subgroup ., In order to demonstrate that the deletion of YFL at the C-terminus of PP2A3 is the causative mutation in imp3 , we performed plasmid rescue ., The PP2A3 gene contains only a single exon , which is predicted to encode a 315 amino acid protein with a predicted molecular weight of 35 , 676 daltons ., An HA-tagged PP2A3 gene with the epitope tag HA immediately following the start codon to avoid compromising the C-terminus and under the regulation of the 637 bp endogenous PP2A3 promoter was transformed into the imp3 mutant strain , and whole cell extract from 6 putative transformants were screened by immunoblotting with an anti-HA antibody ( Table S3 ) ., The HA-PP2A3 protein is predicted to have a molecular weight of 36 , 760 daltons ., The anti-HA antibody recognized a ∼37 kD band in one transformant ( imp3; HA-PP2A3 , Figure 4A ) ., Given the importance of Y313 and L315 ( the equivalent of Y307 and L309 in mammalian cells ) for the function of the PP2A catalytic subunit 33 , we generated various N-terminal HA-tagged mutant forms ( Y313Δ , L315A , L315Δ , and Y313F314L315Δ ) under the same promoter and transformed them into the imp3 cells individually ( Table S3 ) ., All mutant forms of the HA-PP2A3 protein are expressed , as detected by the anti-HA antibody but the amount of the protein is significantly less than the transformed wild-type HA-PP2A3 protein ( Figure 4A ) ., Additionally , we generated a V310T substitution to investigate whether this change has any effect on rescue of the imp3 mutant strain ., The abundance of this protein is about 1 . 5 fold higher than the wild-type HA-PP2A3 transformant ( Figure 4A ) ., An immunoblot with a monoclonal antibody against α-tubulin was used to quantify protein loading ( Figure 4A ) ., Two smaller bands ( ∼27 kD and ∼23 kD ) were also detected by the anti-HA antibody and they may correspond to proteolyzed/truncated PP2A3 ., To ask whether the difference in protein abundance observed in the HA-PP2A3 transformants is due to the abundance of the transgenic HA-PP2A3 transcript or due to protein stability , we measured the transcript levels of PP2A3 by real-time PCR ., In the wild-type strain CC-125 and imp3 , real-time PCR detected only transcript levels of the endogenous PP2A3 transcript ( Figure 4B , blue ) ., In all HA-PP2A3 transformants , levels of two transcripts were detected ., The first primer set detected the combined transcript levels of endogenous PP2A3 and transgenic HA-PP2A3 ( Figure 4B , blue ) ., Overall transcript levels of PP2A3 in all strains tested are comparable ., The second primer set detected only the transcript level of transgenic HA-PP2A3 ( Figure 4B , red ) ., The transcript levels of transgenic HA-PP2A3 in all transformants are about one-quarter of the total PP2A3 transcript levels ., There is no significant difference among different mutant transformants compared to the wild-type HA-PP2A3 transformant ( Figure 4B , red ) ., Therefore , we conclude the differences observed in the protein levels of HA-PP2A3 in different mutant transformants are not due to the abundance of the transgenic HA-PP2A3 transcripts , but rather due to the stability of the HA-PP2A3 proteins ., Transformation of wild-type HA-PP2A3 into imp3 cells successfully rescued the mating defect ( Figure 4C , HA-PP2A3 ) ., Mating between imp3; HA-PP2A3 plus and minus gametes and mating between wild-type and imp3; HA-PP2A3 both produce thick pellicles ( Figure 2A ) ., The V310T change , which is found in 5 independent transformants , partially rescues the mating efficiency to about 60% ( Figure 4C ) ., None of the changes in the YFL motif rescues the mating phenotype ( Figure 4B , Y313Δ , L315A , L315Δ , and YFLΔ ) , which indicates the importance of the last 3 amino acids in the function of PP2A3 during mating ., Thus , we conclude that PP2A3 is encoded by the IMP3 gene and the deletion of nine nucleotides at the C-terminus of this gene causes the defective mating efficiency of imp3 cells ., We further asked whether inhibition of PP2A3 has any effect on mating efficiency ., Okadaic acid ( OA ) , a polyether fatty acid , was shown to inhibit the phosphatase activity of PP2A 40 , enhance phosphorylation of Y307 41 , and inhibit methylation of L309 42 , 43 in vitro ., OA inhibits PP2A at very low concentrations and the dissociation constant ( Ki ) between OA and PP2A is ∼0 . 032 nM 44 ., From in vivo studies , however , the amount of OA required to inhibit PP2A varies from 10 nM in human lung cancer cells 45 to ∼1 µM in MCF7 breast cancer cells 46 ., It is suggested that the entry rate of OA can be affected by pH , temperature , and exposure time to OA 47 ., We tested the effect of OA on Chlamydomonas mating at concentrations of 1 nM , 10 nM , and 1 µM ., The mating efficiency between wild-type CC-124 ( mt− ) and CC-125 ( mt+ ) is around 75% ( Figure 4D ) ., The addition of DMSO and different concentration of OA , for one hour at room temperature , has no significant effect on the mating efficiency ( Figure 4D , blue bars ) ., Similarly , addition of DMSO or OA has no significant effect on the mating efficiency of imp3 mt+×imp3 mt− ( Figure 4D , red bars ) and imp3; HA-PP2A3 mt+×imp3; HA-PP2A3 mt− ( Figure 4D , yellow bars ) ., Pre-treatment of cells with autolysin , an enzyme that removes Chlamydomonas cell walls , before the addition of OA , leads to similar results ( data not shown ) ., In a study on phosphoproteome in Chlamydomonas , cells pre-incubated with 1 . 5 µM OA for 29 hours accumulate 38% more phosphorylated proteins 48 ., Therefore , our OA results indicate that either one hour inoculation is not sufficient for OA to enter Chlamydomonas , or the effect of OA on Chlamydomonas is more complicated than simple inhibition of PP2A3 ., To identify interacting proteins of PP2A3 and to investigate whether changes in the YFL motif lead to changes in protein-protein interactions , we performed immunoprecipitation with the anti-HA antibody ., Two major bands of ∼65 kD and ∼37 kD are obtained by immunoprecipitation from whole cell extract from imp3 gametes transformed with wild-type HA-PP2A3 but not with untransformed imp3 gametes ( Figure 5A ) ., The ∼37 kD band is the HA-PP2A3 , indicated by an immunoblot probed with an anti-HA antibody ( Figure 5A ) ., The ∼65 kD band was excised and subjected to mass spectrometry ., The protein with the most number of peptides ( 94; 27 are unique ) is PP2A-2r ( Cre11 . g477300 ) and it has a predicted size of 64 , 729 daltons ., Changes in Y313Δ , L315Δ , V310T , or YFLΔ did not affect the pull-down of PP2A-2r by the HA antibody ( Figure 5A and S3 ) ., Mass spectrometry of the ∼65 kD band pulled down by HA-PP2A3-V310T and by HA-PP2A3-YFLΔ resulted in 68 ( 27 unique ) and 82 ( 26 unique ) peptides of PP2A-2r , respectively ., Thus , the interaction between the catalytic subunit PP2A3 and the scaffold subunit PP2A-2r is not affected by the changes at the C-terminus of PP2A3 ., PP2A-2r is one of the two PP2A scaffold proteins in the genome and was found in the flagellar proteome 34 , 49 ., Another scaffold subunit FAP14 , which is also found in the flagellar proteome , has a predicted molecular weight of 100 , 787 daltons ., Given that no significant band at ∼100 kD was identified in the immunoprecipitation ( Figure 5A and S3 ) , it is unlikely that PP2A3 interacts with FAP14 ., Since PP2A-2r is present in the flagellar proteome , we asked whether PP2A3 localizes to the flagella ., Flagella were isolated from imp3 and the HA-tagged transformants with the wild-type PP2A3 , V310T , and YFLΔ genes ., The HA-PP2A3 is detected in both wild-type and mutant transformants , but not in the untransformed imp3 flagella with the anti-HA antibody ( Figure 5B ) ., The YFLΔ transgene strain contains only ∼10% of HA-PP2A3 of those found in the wild-type HA-PP2A3 and the V310T strain; this is similar to observations in the whole cell extract immunoblots ( Figure 4A ) ., A smaller band ( ∼29 kD ) , which may represent proteolyzed/truncated HA-PP2A3 , is again recognized by the anti-HA antibody ( Figure 5B ) ., To compare the relative distribution of HA-PP2A3 in Chlamydomonas cell bodies and flagella , we perform immunoblots of HA-PP2A3 on the basis of cell equivalents ( Figure 5C ) ., Protein from equal numbers of whole cells and cell bodies , and from flagella isolated from about 40 times more cells , were used in the analysis ., In both imp3; HA-PP2A3 and imp3; HA-PP2A3 YFLΔ strains , the HA-PP2A3 signal intensity is comparable in all three portions ( Figure 5C ) ., While the flagellar proteins represent less than 5% of protein found in whole cell extract 50 , we do not find a significant enrichment of HA-PP2A3 in the flagella ., In contrast , we observed a significant reduction of the HA-PP2A3 signal in imp3; HA-PP2A3 YFLΔ when compared to imp3; HA-PP2A3 ., Similar to previous observation ( Figure 4A and Figure 5B ) , we noticed additional smaller bands in whole cells , cell bodies , and flagella ., It is intriguing that the smaller bands observed in whole cells/cell bodies and in flagella are different in size and intensity ., It is likely that these represent truncated PP2A3 proteins but the functions of these truncated proteins are unknown ., To ask where the HA-PP2A3 protein localize , we performed immunofluorescence with the HA antibody in six transformant strains ( Figure 6A ) ., In wild-type ( CC-125 ) and untransformed imp3 gametes , there is some non-specific binding of the antibody in the cell body ., In imp3; HA-PP2A3 gametes , robust signals are observed throughout the cells ., In addition , in ∼80% of imp3; HA-PP2A3 gametes , accumulation of the signal is observed around the basal body area ( Figure 6C , blue bars ) ., The same signal intensity and localization is observed in imp3 cells transformed with HA-PP2A3 carrying a V310T mutation ., In comparison , in imp3 cells transformed with the mutant forms of HA-PP2A3 ( Y313Δ , L315A , L315Δ , and YFLΔ ) , the signal intensities of HA are significantly reduced , consistent with what we observed in the immunoblots ( Figure 4A ) ., Less than 10% of these cells showed basal body localization ( Figure 6C ) ., Therefore , we conclude that mutations of the terminal YFL affect the localization of PP2A3 to the basal body region ., Given the accumulation of the HA-PP2A3 proteins in wild-type HA-PP2A3 and V310T cells , we asked whether mating of these gametes with wild-type gametes would lead to change of localization of HA-PP2A3 ( Figure 6B ) ., When Chlamydomonas cells mate , the flagella adhere to each other , leading to cell fusion to form a single cell with four flagella and two nuclei , which is known as a dikaryon ., We examined dikaryons one hour after mixing wild-type gametes ( CC-125×CC-124 ) ; they show a low background of non-specific staining ., In contrast , dikaryons formed between imp3; HA-PP2A3 ( wild-type or V310T ) and wild-type gametes show strong signals throughout the cells ., However , less than 5% of these dikaryons show staining around the basal body area ( Figure 6C , red bars ) ., These results indicate that PP2A3 moves out of the basal body region in dikaryons ., Whole genome sequencing has become an important tool to allow quick identification of causative mutations in Chlamydomonas ( 18 and Dutcher et al . , submitted ) , Caenorhabditis elegans 16 , Drosophila 51 , and humans 52 ., In Drosophila , direct comparison of sequences from parental and EMS mutagenized chromosomes leads to the identification of causative SNPs ., This removes the need for sequence alignment to the reference genome sequences , which eliminates the natural variation of SNPs within different strains 51 ., However , this approach is not feasible to identify mutants whose original strain backgrounds are unavailable ., In C . elegans , a cross to a highly polymorphic strain and whole genome sequencing of a pool of 50 F2 progeny eliminates the need for meiotic mapping ., The number of SNPs drops significantly within a ∼2 Mb region where the mutation resides ., Thus , the number of SNPs of interest is reduced dramatically; it becomes easier to identify the causative mutation 16 ., In the studies of human variants , databases such as dbSNP and The 1000 Genomes Project 53 are available to filter non-causative SNPs/indels ., The filtering results in a reduction of ∼98% of SNPs/indels in a given individual and thus it becomes feasible to identify causative mutations for rare Mendelian diseases 52 ., Similar to the human 1000 Genomes Project , a 1001 Genomes Project on Arabidopsis thaliana was initiated in 2008 19 ., Sequencing of 80 Arabidopsis strains identified ∼5 . 7 million SNPs/indels 54 ., We previously used meiotic mapping to narrow the regions of interest to 269 kb in NG6/fla8-3 and 458 kb in NG30/ift80 , respectively ., Identification of one and six nonsynonymous changes in these regions eventually led to discovery of the causative mutations in these mutant strains 18 ., In an approach similar to that used in C . elegans 16 , we combined a pool of 14 progeny from a cross between a pf27 mutant strain and the highly polymorphic S1C5 strain for whole genome sequencing ., This approach narrows the region of interest to ∼2 Mb on chromosome 12 ( 55 and Alfo

Introduction, Results, Discussion, Materials and Methods

Whole genome sequencing is a powerful tool in the discovery of single nucleotide polymorphisms ( SNPs ) and small insertions/deletions ( indels ) among mutant strains , which simplifies forward genetics approaches ., However , identification of the causative mutation among a large number of non-causative SNPs in a mutant strain remains a big challenge ., In the unicellular biflagellate green alga Chlamydomonas reinhardtii , we generated a SNP/indel library that contains over 2 million polymorphisms from four wild-type strains , one highly polymorphic strain that is frequently used in meiotic mapping , ten mutant strains that have flagellar assembly or motility defects , and one mutant strain , imp3 , which has a mating defect ., A comparison of polymorphisms in the imp3 strain and the other 15 strains allowed us to identify a deletion of the last three amino acids , Y313F314L315 , in a protein phosphatase 2A catalytic subunit ( PP2A3 ) in the imp3 strain ., Introduction of a wild-type HA-tagged PP2A3 rescues the mutant phenotype , but mutant HA-PP2A3 at Y313 or L315 fail to rescue ., Our immunoprecipitation results indicate that the Y313 , L315 , or YFLΔ mutations do not affect the binding of PP2A3 to the scaffold subunit , PP2A-2r ., In contrast , the Y313 , L315 , or YFLΔ mutations affect both the stability and the localization of PP2A3 ., The PP2A3 protein is less abundant in these mutants and fails to accumulate in the basal body area as observed in transformants with either wild-type HA-PP2A3 or a HA-PP2A3 with a V310T change ., The accumulation of HA-PP2A3 in the basal body region disappears in mated dikaryons , which suggests that the localization of PP2A3 may be essential to the mating process ., Overall , our results demonstrate that the terminal YFL tail of PP2A3 is important in the regulation on Chlamydomonas mating .

Whole genome sequencing is a powerful tool to detect changes in genomic DNA ., However , how to identify a causative mutation from over 20 , 000 changes remains a big challenge ., For the unicellular green alga Chlamydomonas , we built a library that consists of over 2 million changes from 16 strains ., A comparison of changes found in a mutant strain with a mating defect , imp3 , to 15 other strains , leads to the identification of a three amino acid deletion in the catalytic subunit of a protein phosphatase 2A ( PP2A3 ) ., The mating defect of imp3 is rescued by an HA-tagged PP2A3 gene ., Introduction of the HA-tagged PP2A3 gene with various mutations in these three amino acids reveals that they play a key role in stabilizing and ensuring the proper localization of PP2A3 ., The ubiquitous enzyme PP2A is involved in diverse cellular processes ., Our discovery that PP2A3 is involved in the Chlamydomonas mating signaling pathway , which also contains the polycystin2 homolog ( PKD2 ) , makes Chlamydomonas mating an excellent model to study ciliary/flagellar signaling ., Since both PP2A and PKD2 play important roles in human health , further investigation of the relationship between these two proteins in Chlamydomonas will facilitate better understanding of their functions .

journal.ppat.1005220

Calcium Regulation of Hemorrhagic Fever Virus Budding: Mechanistic Implications for Host-Oriented Therapeutic Intervention

There is an urgent and unmet need for safe and effective therapeutics against high priority pathogens , including filoviruses ( Ebola and Marburg ) and arenaviruses ( Lassa fever and Junín ) , which can cause fatal infections in humans ., We and others have established that enveloped RNA viruses , including hemorrhagic fever viruses , exhibit a common requirement for host pathways , most notably ESCRT pathway functions , for efficient budding 1–7 ., Indeed as host dependent budding mechanisms are highly conserved within and sometimes across virus families , they represent innovative and immutable antiviral targets for inhibiting virus transmission and disease progression 8–11 ., Importantly , high mutation rates of RNA viruses in general are a factor in their ability to develop resistance to therapeutics that target specific viral proteins or functions 3 , 12–23 ., Consequently , strategies that target specific host mechanisms required by viruses should reduce the development of resistance ., As a number of these host mechanisms , including steps in ESCRT protein function , are targets of calcium regulation , the focus of this study was to determine whether and how hemorrhagic fever viruses mobilize calcium in host cells and whether calcium so mobilized regulates virus budding ., Here we reveal a novel and fundamental requirement for host STIM1- and Orai-mediated Ca2+ entry that regulates late steps of filovirus and arenavirus egress from mammalian cells ., Orai activation is typically linked to either tyrosine kinase or G-protein coupled receptors that activate phospholipase C ( PLC ) and generate diacylglycerol and inositol 1 , 4 , 5-triphoshate ( IP3 ) from membrane phospholipids ., IP3 activates receptor/channels on the endoplasmic reticulum ( ER ) to allow Ca2+ to exit from the ER ., The subsequent drop in ER Ca2+ below the KD ( 400–600μM , 24 ) for the N-terminal EF hands of the ER membrane-resident protein STIM1 initiates a conformational change that promotes STIM1 oligomerization and localization to ER regions adjacent to the plasma membrane ., At the plasma membrane , STIM1 interacts with and activates Calcium-Release Activated Calcium ( CRAC ) channels through which extracellular Ca2+ enters the cell ( reviewed in 25 ) ., CRAC channels are encoded by the Orai family of proteins ( Orai1 , 2 , & 3; 26–28 ) that provide a pathway for sustained extracellular Ca2+ entry to regulate a range of cell functions including gene expression , subcellular trafficking , and the regulation of cell shape and motility 29–31 ., Herein , we demonstrate that both filovirus ( VP40 ) and arenavirus ( Z ) matrix proteins trigger Orai dependent Ca2+ entry in mammalian cells ., In addition , suppression of STIM1 expression and genetic inactivation or pharmacological blockade of Orai inhibits Ebolavirus ( EBOV ) , Marburgvirus ( MARV ) , Lassa Virus ( LASV ) , and Junín Virus ( JUNV ) VLP and infectious virion production and transmission in cell culture ., Together , these data establish a novel and critical role for STIM1- and Orai-mediated Ca2+ entry in late steps of hemorrhagic fever virus egress and establish STIM1 and Orai inhibitors as potential broad-spectrum anti-viral targets for regulation of these and possibly other enveloped RNA viruses that bud by similar mechanisms ., While we previously implicated Ca2+ in EBOV VP40-dependent VLP generation 32 our initial objective here was to understand if and how hemorrhagic fever virus matrix proteins trigger a change in cytosolic calcium in host cells ., To do this we measured intracellular calcium in cells during an extended time course of EBOV and MARV VP40 ( eVP40 and mVP40 , respectively ) and JUNV Z matrix protein-mediated VLP production ., Calcium levels ( R-GECO-1 fluorescence , 33 ) measured in HEK293T cells under physiological conditions for 18–24 hours revealed that eVP40 , mVP40 , and JUNV Z protein expression each induced a time-dependent increase in Ca2+ concentration ( Fig 1 , blue ) , while the GFP-vector backbone induced a negligible Ca2+ increase that plateaued at a low amplitude or declined to baseline levels ( Fig 1 , magenta ) ., To probe the role of Orai1 in these responses we performed identical measurements in an HEK293T line that stably expresses a dominant negative mutant Orai1 having a glutamic acid ( E ) to alanine ( A ) substitution in its ion selectivity filter ( E106A ) ., Incorporation of even a single Orai1 E106A subunit into endogenous WT Orai channels exerts a dominant negative block of its Ca2+ permeation 34 ., Importantly , both WT and E106A Orai HEK293T cells exhibited a similar transient Ca2+ elevation following treatment with the membrane permeant SERCA pump inhibitor thapsigargin in Ca2+ free bath solution , indicating that ER stores were intact in E106A Orai1 expressing HEK293T ( S1 Fig ) ., The absence of a secondary increase in cytoplasmic Ca2+ ( S1 Fig , left panel ) following reperfusion of HEK293T Orai1 E106A cells with Ca2+-containing Ringers solution ( S1 Fig , right panel ) verified the Orai permeation defect of this line ., Significantly , in permeation defective Orai1 E106A cells neither eVP40 , mVP40 , JUNV Z protein ( Fig 1 , yellow ) , nor GFP vector ( Fig 1 , orange ) triggered any change in cytoplasmic Ca2+ levels indicating that Ca2+ elevations initiated by expression of EBOV , MARV , and JUNV matrix proteins required and resulted from Ca2+ entry though Orai channels ., Consistent with these results from Orai E106A HEK293T cells and specifically the role of Orai , the Orai inhibitor Synta66 similarly blocked the eVP40-mediated increase in cytoplasmic Ca2+ ( Fig 1 , lower right panel ) in WT HEK293T cells ., Given the ability of EBOV , MARV , and JUNV matrix proteins to initiate an Orai-dependent Ca2+ signal in HEK293T cells , we assessed the role of Orai1-mediated calcium signals in eVP40 , mVP40 , LASV Z or JUNV Z mediated VLP production in WT and Orai1 E106A HEK293T cells ., Consistent with a role for Ca2+ entry via Orai1 in VLP production , we found that Orai1 E106A cells did not support efficient filovirus or arenavirus VLP production ( Fig 2 ) ., Indeed , levels of eVP40 VLPs from Orai1 E106A cells were ~50 fold lower than that from WT cells ( Fig 2A , VLPs ) ., Similarly , production of mVP40 VLPs exhibited an even greater dependence on Orai1-mediated calcium entry ( Fig 2B , VLPs ) , as mVP40 VLPs from Orai1 E106A HEK293T cells were ~100 fold lower than that from WT cells ( Fig 2B ) ., Orai similarly regulated JUNV Z ( Fig 2C ) and LASV Z ( Fig 2D ) VLP production as both JUNV Z and LASV Z protein-mediated VLP production from E106A cells was ~100 fold lower than that from WT HEK293T cells ., In all instances , cellular levels of VP40 or Z were similar in WT and E106A cells , indicating no general requirement for Orai1-mediated Ca2+ entry in viral protein expression ( Fig 2A–2D; Cells ) ., Together , these data point to a conserved and selective role for Orai-mediated Ca2+ entry in hemorrhagic fever virus budding ., Implicit in this common critical role for Orai1-mediated Ca2+ entry in EBOV , MARV , JUNV , and LASV VLP production is an upstream requirement for STIM1 , the only known trigger for Orai activation in mammalian cells ., STIM1 is a single pass ER membrane protein whose activity is regulated by ER Ca2+ binding ., Ca2+ dissociation from STIM1 following a decrease in ER concentration triggers a N-terminal conformational change that initiates its multimerization and relocalization within the ER membrane to domains juxtaposed to the plasma membrane 35–37 ., The resulting subplasmalemmal STIM1 clusters physically activate Orai channels to allow extracellular Ca2+ to enter the cell 25 ., Using eVP40 VLP budding as our model , we probed the role of STIM1 in VLP formation by assessing VLP production from STIM1 suppressed HEK293T cells ., eVP40 VLP budding from STIM1 suppressed cells was reduced by approximately 10 fold relative to that from cells receiving random siRNAs or no siRNA ( Fig 3A ) , and the loss of STIM1 had no effect on cellular expression of eVP40 or actin ( Fig 3A; Cells ) ., To further confirm this requirement for STIM1 in VLP formation , we utilized a bicistronic vector to suppress endogenous STIM1 ( by targeting the 5’ UTR ) and rescued its expression with exogenous human STIM1 translated from a shRNA resistant cDNA ( shSTIM1-STIM1 plasmid ) ( Fig 3B ) ., HEK293T WT cells expressing a fixed amount of eVP40 were transfected with increasing amounts of the shSTIM1-suppression vector or empty vector ( Fig 3B ) ., While cellular eVP40 expression levels were equivalent under all conditions ( Fig 3B , Cells ) , progressive suppression of STIM1 expression led to a dose-dependent decrease in eVP40 VLP production ( Fig 3B , VLPs , middle panel ) ., Importantly , STIM1 re-expression in suppressed cells fully rescued eVP40 VLP production across all levels of STIM1 suppression ( Fig 3B , VLPs , bottom panel ) ., Similar to results with STIM1 shRNA , budding of eVP40 VLPs was significantly reduced ( Fig 3C , VLP ) following siRNA mediated STIM1 suppression ( >90% , Fig 3C , middle panel ) ; and over-expression of exogenous STIM1 restored eVP40 VLP production ( Fig 3C ) ., Taken together with results from experiments performed on E106A HEK293T cells , these data definitively establish a role for STIM1/Orai dependent Ca2+ signals in regulation of VLP egress ., Genetic approaches outlined above to modulate STIM1 expression and Orai1 permeation establish a novel and common critical role for STIM1 and Orai1 in filovirus and arenavirus budding ., Given the broad utility of targeting ion channels to regulate a range of cell physiological functions , we asked whether pharmacological blockade of Orai might represent an effective strategy for regulating filovirus and arenavirus budding ., Although high affinity Orai1 blockers for in vivo applications are not available at present , we tested several commercially available inhibitors including Synta66 and 2-APB , both of which inhibit Orai-mediated Ca2+ entry in HEK293T cells at micromolar levels ( 10–50 μM ) 38 , 39 without impacting calcium release from the ER ( S2 Fig ) ., Both 2-APB and Synta66 inhibited eVP40- ( Fig 4A and 4B ) and mVP40-induced ( Fig 4C and 4D ) VLP production with identical potency as inhibition of Orai-mediated calcium entry , and neither drug affected cellular expression of VP40 or actin ( Fig 4A–4D ) ., A concentration of 2-APB that fully blocks Orai1 channels ( 50 μM ) 38 inhibited eVP40 VLP production ( Fig 4A , right panel ) by ~5 fold and mVP40 VLP production ( Fig 4C , right panel ) by ~50 fold ., Likewise , Synta66 ( 50μM ) substantially inhibited eVP40 ( ~5 fold ) and mVP40 VLP ( ~10 fold ) production ( Fig 4B and 4D ) with no effect on steady state levels of cellular VP40 or actin and without altering membrane localization of viral proteins ( S3 Fig ) ., Importantly , as neither 2-APB ( Fig 4E ) nor Synta66 ( Fig 4F ) exerted cytotoxic effects on cells under conditions of these measurements ( cell viability , cellular production of VP40 , or VP40 membrane localization , Fig 4 and S3 Fig ) , their anti-budding activity can be attributed to inhibition of Orai-mediated Ca2+ entry ., Finally , an additional Orai selective inhibitor ( RO2959 40 ) , which recently became commercially available , also blocks eVP40 VLP formation ( ~10-fold ) with a potency that parallels its inhibition of calcium permeation of the channel ( S4 Fig , IC50 = ~2 . 5μM ) ., Thus , the sensitivity of budding to three chemically distinct Orai inhibitors , at the same concentration that blocks calcium permeation of Orai , further substantiates the critical role of Orai-mediated Ca2+ entry in VLP production ., We next sought to validate VLP findings by examining the effect of the Orai1 inhibitors Synta66 and 2-APB on budding of the live attenuated Candid-1 JUNV vaccine strain 41 , 42 ., Briefly , VeroE6 cells infected with live attenuated Candid-1 JUNV were cultured in the absence or presence of Orai inhibitors , and infectious virions produced from these cells were quantified in a focus forming assay ( Fig 5 ) ., Enumeration of JUNV foci revealed a statistically significant , dose-dependent reduction in JUNV virus production following treatment with Synta66 ( Fig 5A ) or 2-APB ( Fig 5B ) ., Moreover , neither compound affected the viability of cells cultured under conditions mimicking those used for JUNV infection experiments ( Fig 5A and 5B , right panels ) , nor affected the synthesis of JUNV GP in infected VeroE6 cells at any concentration tested ( Fig 5A and 5B , Western blots ) ., Together , these findings corroborate results of VLP budding assays ( Fig, 2 ) and demonstrate that Orai1-mediated calcium entry is required for efficient budding of infectious JUNV ., Based on the general requirement we identify for Orai channels in filovirus and arenavirus VLP production and JUNV ( Candid-1 ) budding , we next sought to determine whether Orai channels regulate spread of infectious pathogenic strains of EBOV , MARV , LASV , and JUNV ., We first examined viral spread , an indicator of efficient viral budding , in HEK293T cells that constitutively express the dominant negative permeation defective variant of Orai1 ( E106A ) used in VLP assays described above ( Fig 2 ) ., These cells were infected at a low multiplicity of infection ( MOI ) , which resulted in the infection of approximately 2–5% of the cells ., Cells were then incubated for a period of time that equates to several rounds of viral replication , allowing us to assess viral spread ., We observed that the percent of Orai1 E106A expressing cells infected with live BSL-4 variants of EBOV , MARV , JUNV , or LASV was significantly lower than Orai WT cells infected with the same viruses ( Fig 6 ) ., These results are consistent with a role for Orai in the spread of infectious filoviruses and arenaviruses ., We next assessed the effect of the Orai blocker Synta66 on the spread of these BSL-4 pathogens , because it is a more consistent Orai blocker than 2-APB ., Viral spread was assessed by infecting HeLa cells with LASV , JUNV , MARV , or EBOV at a low MOI and then treating with vehicle or Synta66 at the indicated concentrations beginning 1 hour post infection and for the duration of experiments ., Seventy two ( LASV , JUNV ) or 96 ( MARV , EBOV ) hours post infection we quantified the percentage of cells infected with virus ., For each virus , we observed a significant Synta66 dose-dependent decrease in the percentage of cells infected ( Fig 7A ) ., Consistent with inhibition of viral spread , we also observed a general decrease in the number and size of infected cell clusters with increasing Synta66 concentration ( Fig 7B ) ., Similar to the more potent inhibition by Synta66 of mVP40- versus eVP40-mediated VLP production ( Fig 4 ) , Synta66 also exerted more potent inhibition of live MARV than EBOV ., Interestingly , the spread of both arenaviruses was more sensitive to Orai inhibition than either filovirus ( Fig 7A ) ., In general , cultures treated with higher concentrations of Synta66 and for a prolonged period of time ( 72–96 hours ) contained fewer cells than vehicle control treated cultures as measured by the number of nuclei ( Fig 7B ) ., For this reason , we normalized the results as the percent of cells infected at the time of analysis for each condition ., The decrease in cell numbers , however , does not appear to reflect toxicity ( see Figs 4F and 5A ) ., Indeed , as HeLa cells autonomously divide , fewer cells more likely reflects an effect of prolonged Synta66 treatment on cell proliferation ., Nonetheless , we evaluated Synta66 induced toxicity by two separate methodologies ., Cell-titer Glo “viability” measurements revealed that prolonged Synta66 produced a dose-dependent decrease in ATP ( S5 Fig ) that is attributed to a decrease in the overall number of cells in cultures and not an effect of Synta66 on cell viability ( see Fig 7B ) ., We then utilized an Alamar Blue assay to assess the metabolic health of Synta66 treated cells ., Indeed , cellular oxidation-reduction potential of Synta66 treated and vehicle control treated cells were equivalent , confirming comparable metabolic activity ( S5 Fig ) ., Thus , while prolonged Synta66 treatment resulted in lower overall cell numbers , those cells that are present are metabolically healthy and are fully capable of producing virus ., We next sought to definitively establish that the effect of Synta66 on virus spread is due to inhibition of virus egress and not entry ., We first pretreated HeLa cells with Synta66 and then infected with a high MOI of LASV , JUNV , MARV , or EBOV ., Cells were then fixed after only 2–3 viral replication cycles ., Infecting with a high MOI and fixing soon after infection ensured that the extent of infection minimally involves spread between cells and rather reflects the extent of primary infection ., Under these high MOI conditions , we observed relatively little effect of Synta66 on infection levels with only modest inhibition of infection evident at high Synta66 concentrations ( S6 Fig ) ., Further confirmation that Synta66 blocks egress of live filoviruses and arenaviruses was obtained by assessing the amount of virus released into culture supernatants ., Supernatants were collected between 48 and 96 hours post-infection with JUNV , LASV , MARV , or EBOV from Synta66 or vehicle treated HeLa cells ., Consistent with all of our VLP ( Figs 2–4 ) and live virus data ( Figs 5 , 6 , 7A and 7B ) we found that Synta66 ( 30μM ) significantly reduced the titer of infectious Lassa , Junín , Marburg , and Ebola virion particles in culture supernatants ( Fig 8 ) ., Taken together , this data provide a clear and comprehensive demonstration that Synta66 treatment significantly impairs authentic filovirus and arenavirus budding and release from infected cells ., In summary , our results clearly establish that, 1 ) Orai1-mediated Ca2+ entry is a critical virus-triggered host signal that regulates filovirus and arenavirus budding , and, 2 ) STIM1 and Orai1 represent novel targets for broad-spectrum control of these emerging and often fatal viruses ., Indeed , the conserved role for Orai mediated calcium entry among these four hemorrhagic fever viruses raises the interesting possibility that Orai inhibitors may have general utility for broad spectrum control of these and other enveloped RNA viruses that bud by similar Ca2+ dependent mechanisms ., The recent catastrophic outbreak of EBOV in West Africa highlights the need to develop therapeutics for EBOV and other hemorrhagic fever viruses ., Indeed , much progress has been made toward the development of candidate vaccines and therapies against EBOV that are currently in clinical trials ., Nevertheless , it is critically important that we improve our understanding of the mechanisms of hemorrhagic fever virus pathogenesis not only to identify novel viral targets , but also to identify host targets and common mechanisms that these viruses require for completion of their life cycles as these could lead to the development of broad spectrum host oriented therapeutics ., A key advantage of therapeutics that target conserved host pathways required broadly by families of viruses for transmission is the potential for broad spectrum efficacy compared with drugs that target strain specific viral targets ., Moreover , host targets should be essentially immutable and thereby insensitive to selective pressures that normally allow pathogens to develop drug resistance 3 , 12–23 ., Here we focused on the second messenger Ca2+ and the host proteins responsible for its mobilization and asked whether calcium signals within host cells orchestrate virus assembly and budding ., While calcium has been implicated generally in EBOV and HIV-1 budding 32 , 43 , 44 , previous efforts have not addressed if and how matrix proteins encoded by filoviruses or arenaviruses might trigger changes in Ca2+ concentration in host cells ., Herein , we demonstrate for the first time that the filovirus matrix protein VP40 and JUNV Z protein trigger STIM1/Orai activation and that the resulting influx of extracellular Ca2+ controls both VLP formation and production of infectious filovirus and arenavirus progeny ., Moreover , using Orai channel inhibitors , Orai permeation defective lines , and by suppressing STIM1 expression , we establish STIM1 and Orai as effective host targets for pharmacological regulation of virus egress ., It should be noted; however , that we cannot rule out a role for other Orai isoforms ( Orai2 and Orai3 ) in to the residual budding observed for live virus from E106A or Synta66 treated cells ., While we have established a critical role for Orai-mediated calcium entry in budding of hemorrhagic fever viruses , the mechanism by which Ca2+ does so remains an important question and the focus of ongoing efforts ., Indeed , a number of critical steps implicated in efficient budding of enveloped RNA viruses have been linked to cellular Ca2+ signals , including the activation and localization of specific ESCRT components ., Although not absolutely required , the ESCRT pathway has been shown to play a key role in efficient budding of a plethora of RNA viruses including filoviruses , arenaviruses , rhabdoviruses , and retroviruses 5 ., It is tempting to speculate that the observed calcium regulation of budding described here may be linked mechanistically to the role of ESCRT during virus egress ., For example , the structure , activation , and interactions of ESCRT-related proteins such as Tsg101 , Nedd4 , and Alix have been shown to be regulated in part by calcium 44–47 ., Additionally , given that Ca2+ control of membrane repair reflects ESCRT induced shedding of damaged membrane 48 , one might also speculate that Ca2+ dependent mechanisms are similarly triggered by insertion of viral proteins in the plasma membrane ., Studies underway are thus focused on determining whether Ca2+ controls budding through regulation of ESCRT pathway function ., STIM1 and Orai1 mediated Ca2+ signals have been implicated in distinct steps of the life cycle of other viruses including the replication of Rotaviruses , which are non-enveloped RNA viruses that do not bud from the plasma membrane ., Constitutive STIM1 ( and Orai1 ) activation observed in rotavirus-infected cells reflects an effect of its nonstructural protein 4 ( NSP4 ) on endoplasmic reticulum Ca2+ permeability 49 ., Indeed , ongoing efforts within our group to understand the mechanisms by which hemorrhagic fever virus matrix proteins trigger STIM1/Orai activation include testing whether VP40 might likewise trigger Ca2+ leak from the ER by inhibiting SERCA pump activity ., Furthermore , Ca2+ influx also seems to regulate entry of West Nile virus 50 , 51 , Coxsackievirus 52 , 53 , Hepatitis B virus 54 , and Epstein Barr virus 55 , 56 ., Recently it was shown that subunits of a functionally distinct family of voltage-gated calcium channels ( VDCCs ) also play a role in JUNV and Mouse Mammary Tumor pseudovirus entry and infection 57 and that the VDCC blockers nifedipine and verapamil suppressed host cell entry by these viruses ., Surprisingly; however , in this instance the involvement of VDCC subunits seemed to be distinct from any role in regulating Ca2+ levels ., How VDCC inhibitors might operate independently of any action on VDCC Ca2+ permeation is unclear , but could reflect the promiscuous affinity of VDCC inhibitors for channels including voltage-gated potassium ( Kv ) channels ., Indeed , verapamil and nifedipine also block voltage-gated potassium channels that set the membrane potential of non-excitable cells 58 , 59 ., Depolarization of the plasma membrane as a result of Kv channel blockade could indirectly block calcium entry by dissipating the electrical driving force ( membrane potential ) required for calcium permeation of Orai 60–62 ., While these studies cumulatively point to additional roles for Orai1-mediated and independent Ca2+ influx in steps of infection and replication used by a range of disparate viruses , these roles are distinct from the selective requirement we identify for Orai-dependent calcium entry in budding of filoviruses and arenaviruses ., However , Orai might represent a conserved target for regulating budding of additional enveloped RNA viruses , including retroviruses such as HIV-1 , which buds by similar mechanisms ., Indeed , similar to hemorrhagic fever viruses , the HIV-1 matrix protein Gag directs HIV-1 budding in part , via well-characterized L-domain interactions with ESCRT proteins , and Gag mediated VLP formation also exhibits dependence on Ca2+ regulation 43 ., Further study is needed to fully assess the role for calcium in the HIV-1 lifecycle because , unlike filoviruses and arenaviruses , Gag-mediated VLP production was found to be insensitive to high concentrations of 2-APB ( up to 200uM ) that fully block Ca2+ permeation of Orai channels 44 ., In conclusion , we provide the first direct evidence that host Ca2+ signals , triggered by virus activation of STIM1 and Orai , are among key host mechanisms that orchestrate late steps of filovirus and arenavirus assembly and budding ., Importantly , from a therapeutic perspective , Orai channels are ubiquitously expressed and like ion channels in general , they represent pharmacologically accessible ( cell surface ) therapeutic targets ., While Orai1 inhibitors by themselves appear to have broad spectrum efficacy , an exciting possibility raised by our results is that drug cocktails formulated to target sequential steps in the virus life cycle , including entry , L-domain/host interactions , and other steps involved in budding , could produce enhanced potency , coverage and efficacy over approaches targeting any one host dependent step in the virus life cycle ., Thus , while other calcium channel modulators identified may have distinct targets and even calcium independent effects , they may synergize with Orai1 , and also L-domain inhibitors we’ve described previously that block VP40 and Z protein L-domain interactions with host Nedd4 and Tsg101 42 , 63 ., Finally , the ability of certain individuals to survive hemorrhagic fever virus infection seems to reflect their capacity to mount a robust anti-viral immune response ., In the context of the severity and the acute nature of these viral diseases , the impact of side effects and even minor effects on cell proliferation that might be associated with long term administration of Orai inhibitors that would be required for immune suppression and immune modulation , may be tolerable in the context of infection with these highly pathogenic and often fatal viruses ., Indeed , there is no evidence from murine models that the loss of STIM or Orai activity or function would affect antigen induced lymphocyte activation required for an antiviral immune response 64 , 65 ., Thus our prediction is that administration of Orai1 or STIM1 inhibitors , or cocktails that could also include L-domain inhibitors , would slow or dampen virus transmission within and between individuals , and thereby could provide infected individuals additional time needed to mount a protective adaptive immune response ., Although Synta66 and the more potent compound RO2959 are no longer being developed as therapeutics , several smaller pharmaceutical companies and academic groups persist in efforts to develop potent Orai1 inhibitors to suppress the pathogenesis of chronic immune-mediated and inflammatory diseases ., If and when these become available , direct inhibition of enveloped RNA virus budding from host cells and transmission between individuals may represent an entirely novel use for these channel blockers ., HEK293T , HeLa , and VeroE6 cells were maintained in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% fetal calf serum ( FCS ) , penicillin ( 100 U/ml ) /streptomycin ( 100μg/ml ) at 37°C in a humidified 5% CO2 incubator ., The stable HEK293T Orai1 E106A mutant-expressing cell line was maintained in DMEM with 10% FCS , penicillin ( 100 U/ml ) /streptomycin ( 100μg/ml ) in the presence of 500 μg/ml of G418 ., HeLa cells were maintained in Minimal Essential Medium ( MEM ) with 5% fetal bovine serum , penicillin ( 100 U/ml ) /streptomycin ( 100μg/ml ) at 37°C in a humidified 5% CO2 incubator ., The pCAGGS based plasmids expressing EBOV VP40 , MARV VP40 , JUNV Z , LASV Z , and GFP-eVP40 have been described previously 42 , 63 , 66 ., mVP40 and JUNZ Z protein are flag tagged while eVP40 was detected using an anti-eVP40 polyclonal antibody previously described 67 ., mVP40 and JUNV Z protein were detected with an anti-flag monoclonal antibody ( Sigma-Aldrich ) , and STIM1 was detected with a rabbit anti-STIM1 specific polyclonal antibody ( gift of Dan Billadeau , Mayo Clinic ) ., 2-aminoethoxy diphenyl borate ( 2-APB , Sigma Aldrich ) , Synta66 , and RO2959 ( Glixx Labs , Southborough , MA ) were freshly prepared from stock solutions in DMSO ., Cell viability in VLP budding and live virus infection assays was examined using an MTT assay ( Amresco ) ., 5×103 HEK293T or VeroE6 cells were plated in collagen-coated 96-well tissue culture plates in triplicate ., Cells were transfected with empty vector using Lipofectamine for 6 hours , and incubated in serum-free or 2% FCS in phenol-red-free OPTI-MEM in the presence of Synta66 or 2-APB at the indicated concentrations for 20 hours , which mimics the transfection and treatment conditions for VP40 VLP budding ., 20μl of MTT solution ( 5mg/ml in PBS ) was added into each well and cells were incubated for 3 . 5 hours ., Media was discarded and 150 μl DMSO was added ., Absorbance was determined by spectrophotometry using a wavelength of 590 nm ., For experiments with BSL-4 variants of filoviruses and arenaviruses , HeLa cells were seeded in 96 well plates ~24 hours prior to addition of Synta66 or vehicle control at indicated concentrations ., Cells were incubated for 72 or 96 hour at 37°C in a humidified 5% CO2 incubator and viability was assessed using either CellTiter-Glo assay ( Promega ) , or AlamarBlue assay ( Life Technologies ) , in accordance with manufacturer’s instructions ., HEK 293T Orai1-wild type and Orai1-E106A mutant cells ( kind gift from Dr . Jonathan Soboloff , Temple University ) were plated at 5x105 cells/well in a 2-chambered Lab-Tek II Chambered #1 . 5 slide ( Nunc , Rochester , NY ) and grown overnight at 37°C , 5% CO2 in Dulbecco’s modified Eagle’s Medium supplemented with 4 . 5g/L glucose , L-glutamine , sodium pyruvate ( Mediatech , Inc . , Manassas , VA ) , 10% FBS ( Gibco ) , and 1% Penicillin/Streptomycin ( Gibco ) ., Cells were transfected with 1μg of R-GECO-1 plasmid ( Addgene , Cambridge , MA ) using Lipofectamine 2000 reagent ( Invitrogen ) according to manufacturer’s instructions in phenol-red-free OPTI-MEM ., The next day , cells were transfected with GFP-eVP40 fusion plasmid ( GFP-eVP40 ) using Lipofectamine 2000 reagent ( Invitrogen ) according to manufacturer’s instructions in phenol-red-free OPTI-MEM ., Six hours post transfection , fresh phenol-red-free OPTI-MEM was added to the wells , and cells were imaged at 37°C and 5% CO2 in a custom environmental chamber for the duration of the imaging on a Leica DMI4000 with Yokagawa CSU-X1 Spinning Disk Microscope with a 20X dry objective ., Cellular R-GECO-1 fluorescence was imaged every 4 seconds for 1 minute periods , repeated every 10 minutes over 18 hours with a Hamamatsu 16-bit cooled EMCCD camera ., Imaging and data analysis were performed using the Metamorph 7 . 6 imaging suite ., Normalized fluorescence intensity ( F/F0 , where F0 is calculated as the average fluorescence intensity for the initial 1 minute interval ) was calculated for each region of interest ( ROI ) in the time-series ., WT HEK293T or HEK293T E106A cells were seeded in collagen-coated six-well plates and transfected with 0 . 5μg of the indicated expression plasmids using Lipofectamine ( Invitrogen ) and the protocol of the supplier ., At 6 hours post-transfection , cells were incubated in serum-free OPTI-MEM media or 2% FCS DMEM for 20–24 hours ., Cells were incubated with vehicle ( DMSO ) a

Introduction, Results, Discussion, Materials and Methods

Hemorrhagic fever viruses , including the filoviruses ( Ebola and Marburg ) and arenaviruses ( Lassa and Junín viruses ) , are serious human pathogens for which there are currently no FDA approved therapeutics or vaccines ., Importantly , transmission of these viruses , and specifically late steps of budding , critically depend upon host cell machinery ., Consequently , strategies which target these mechanisms represent potential targets for broad spectrum host oriented therapeutics ., An important cellular signal implicated previously in EBOV budding is calcium ., Indeed , host cell calcium signals are increasingly being recognized to play a role in steps of entry , replication , and transmission for a range of viruses , but if and how filoviruses and arenaviruses mobilize calcium and the precise stage of virus transmission regulated by calcium have not been defined ., Here we demonstrate that expression of matrix proteins from both filoviruses and arenaviruses triggers an increase in host cytoplasmic Ca2+ concentration by a mechanism that requires host Orai1 channels ., Furthermore , we demonstrate that Orai1 regulates both VLP and infectious filovirus and arenavirus production and spread ., Notably , suppression of the protein that triggers Orai activation ( Stromal Interaction Molecule 1 , STIM1 ) and genetic inactivation or pharmacological blockade of Orai1 channels inhibits VLP and infectious virus egress ., These findings are highly significant as they expand our understanding of host mechanisms that may broadly control enveloped RNA virus budding , and they establish Orai and STIM1 as novel targets for broad-spectrum host-oriented therapeutics to combat these emerging BSL-4 pathogens and potentially other enveloped RNA viruses that bud via similar mechanisms .

Filoviruses ( Ebola and Marburg viruses ) and arenaviruses ( Lassa and Junín viruses ) are high-priority pathogens that hijack host proteins and pathways to complete their replication cycles and spread from cell to cell ., Here we provide genetic and pharmacological evidence to demonstrate that the host calcium channel protein Orai1 and ER calcium sensor protein STIM1 regulate efficient budding and spread of BSL-4 pathogens Ebola , Marburg , Lassa , and Junín viruses ., Our findings are of broad significance as they provide new mechanistic insight into fundamental , immutable , and conserved mechanisms of hemorrhagic fever virus pathogenesis ., Moreover , this strategy of targeting highly conserved host cellular protein ( s ) and mechanisms required by these viruses to complete their life cycle should elicit minimal drug resistance .

journal.pgen.1005094

Genome-Wide Association Study Identifies Nox3 as a Critical Gene for Susceptibility to Noise-Induced Hearing Loss

Noise-induced hearing loss ( NIHL ) is a worldwide leading occupational health risk in industrialized countries and is the second most common form of sensorineural hearing impairment , after presbyacusis 1 ., In the United States , roughly 10% of the total population is exposed daily to hazardous levels of noise in the workplace 2 ., The most extreme workplace environment for NIHL is the Armed Forces ., According to the Department of Veterans Affairs , hearing loss is the most common disability among U . S . troops in the Middle East ., The financial impact of these disability claims on the VA is staggering and likely will continue to grow ., According to the American Tinnitus Association ( http://www . ata . org/ ) , the number of disability claims from hearing injury is expected to increase by 18% per year with a total cost of $1 . 2 billion annually 3 ., Risk could be reduced with a better understanding of the biological processes that modulate susceptibility to damaging noise ., It is believed that NIHL is a complex disease resulting from the interaction between environmental and genetic factors and it is well recognized that people with similar exposures to noise show variation in the amount of hearing loss , indicative of a genetic component 4 ., Twin studies estimate heritability for noise-induced hearing loss ( NIHL ) of approximately 36% 5 ., The discovery of gene by environment interactions in human disease , such as susceptibility to NIHL , has many inherent difficulties , most notably , controlling for exposure ., Although several candidate gene association studies for NIHL in humans have been conducted , each is underpowered , un-replicated , and accounts for only a fraction of the genetic risk ., In addition , no heritability studies have been performed , since families , where all subjects are exposed to identical noise conditions , are almost impossible to collect ., The genetic basis of NIHL has been clearly demonstrated in animals as different susceptibilities to noise have been seen in different inbred stains of mice 4 ., Mouse strains ( C57BL/6J ) exhibiting age-related hearing loss ( AHL ) were shown to be more susceptible to noise than other strains 6 ., Also , several knockout mice including SOD1-/- 7 , GPX1-/- 8 , PMCA2-/- 9 and CDH23+/- 10 were shown to be more sensitive to noise than their wild-type littermates ., The mouse has been an essential animal model for studies in hearing loss , and advances in mouse genetics , including genome sequence and high density single-nucleotide polymorphism ( SNP ) maps , provide a suitable system for the study of a complex trait such as NIHL 6 ., The identification of novel genes is crucial for the discovery of new pathways and gene networks that will improve our knowledge of basic hearing biology and identify new therapeutic targets with the potential to combat NIHL ., Due to the limitations of human genome-wide association study ( GWAS ) and quantitative trait locus ( QTL ) analyses in mice , we have chosen to use a genome-wide association strategy incorporating the Hybrid Mouse Diversity Panel ( HMDP ) ., The HMDP is a collection of classical inbred ( CI ) and recombinant inbred ( RI ) strains whose genomes have been sequenced and/or genotyped at high resolution 11 ., Power calculations have demonstrated that this panel is superior to traditional linkage analysis and is capable of detecting loci responsible for 5% of the overall variance ., Several studies have successfully mapped candidate loci for complex traits using this panel and we have recently published a meta-analysis for age-related hearing loss incorporating the HMDP 12 13 14 15 ., In this manuscript we describe , for the first time , an association analysis with correction for population structure in the mapping of several loci for susceptibility to NIHL in inbred strains of mice ., After completing a preliminary screen of the HMDP , an intriguing locus appeared warranting further exploration ., Herein , we describe a genome-wide significant peak on ( Chr . ) 17 within a haplotype block containing NADPH oxidase-3 ( Nox3 ) and provide evidence supporting its role in susceptibility to NIHL ., Furthermore , we demonstrate frequency-specific genetic susceptibility within the mouse cochlea ., The Institutional Care and Use Committee ( IACUC ) at University of Southern California , Los Angeles , approved the animal protocol for the HMDP strains and the Nox3het mice ( IACUC 12033 ) ., HMDP strains and C57BL/6JEiJ Nox3het ( Nox3het/Nox3het , Nox3het/+ and wild-type ) were anesthetized with an intraperitoneal injection of a mixture of ketamine ( 80 mg/kg body weight ) and xylazine ( 16 mg/kg body weight ) ., A detailed description of the HMDP ( strain selection , statistical power and mapping resolution ) is provided in Bennett BJ , et al . 2010 ., 11 ., Approximately four female mice for each HMDP strain were purchased from the Jackson Laboratory ( Bar Harbor , ME ) ., Only female mice were tested to avoid confounding effects of sex ., Mice were 4 weeks of age , and to ensure adequate acclimatization to a common environment , mice were aged until 5 weeks ., 5-week-old mice were selected to eliminate the potential effects of age-related hearing loss contributing to our phenotype ., All mice were maintained on a chow diet until sacrifice ., Common and recombinant inbred strains were previously genotyped by the Broad Institute ( www . mousehapmap . org ) ., Of the 140 , 000 SNPs available , 108 , 064 were informative ( allele frequency ≥ 5% and less than 20% missing data ) and were used for the association analysis ., Stainless-steel electrodes were placed subcutaneously at the vertex of the head and the right mastoid , with a ground electrode at the base of the tail ., Body temperature was maintained and monitored ., Artificial tear ointment was applied to the eyes ., Each mouse was recovered on a heating pad at body temperature ., Auditory signals were presented as tone pips with a rise and a fall time of 0 . 5 msec and a total duration of 5 msec at the frequencies 4 , 8 , 12 , 16 , 24 , and 32 kHz ., Tone pips were delivered below threshold and then increased in 5 dB increments until goal of 100 dB ., Signals were presented at a rate of 30/second ., Responses were filtered with a 0 . 3 to 3 kHz pass-band ( x10 , 000 times ) ., For each stimulus intensity 512 waveforms were averaged ., Hearing threshold was determined by inspection of auditory brainstem response ( ABR ) waveforms and was defined as the minimum intensity at which wave 1 could be distinguished ., Data was stored for offline analysis of peak-to-peak ( P1-N1 ) values for wave 1 amplitudes ., Post-exposure thresholds were evaluated by the same method 2 weeks post-exposure ., Distortion product otoacoustic emissions ( DPOAEs ) were analyzed as input/output ( I-O ) functions with 2f1- f2 ( primary measure ) ., Primary tones were set at a ratio of f2/f1 = 1 . 2 with the f2 between 8 to 32 kHz ( f2 level set 10 dB less than the f1 level ) and L2 ranging from 20 to 70 dB ., The noise floor was measured by averaging 6 spectral points ( above and below the 2f1- f2 ) ., After both waveform and spectral averaging DPOAEs were extracted ., Threshold was defined as the L2 level needed to produce a DPOAE of 0 dB SPL with a signal to noise ratio ( SNR ) ≥ 3 dB ., 6 week-old mice were exposed for 2 hours to 10 kHz octave band noise ( OBN ) at 108 dB SPL using a method adapted from Kujawa and Liberman ( 2009 ) 16 ., The OBN noise exposure was previously described 17 ., For 2 hours , mice were placed in a circular ¼-inch wire-mesh exposure cage with four shaped compartments and were able to move about within the compartment ., The cage was placed in a MAC-1 soundproof chamber designed by Industrial Acoustics ( IAC , Bronx , NY ) and the sound chamber was lined with soundproofing acoustical foam to minimize reflections ., Noise recordings were played with a Fostex FT17H Tweeter Speaker built into the top of the sound chamber ., Calibration of the damaging noise was done with a B&K sound level meter with a variation of 1 . 5 dB across the cage ., A data acquisition board from National Instruments ( National Instruments Corporation , Austin , Texas ) was regulated by custom software ( used to generate the stimuli and to process the responses ) ., Stimuli were provided by a custom acoustic system , made up of two miniature speakers , and sound pressure was measured by a condenser microphone ., Testing involved the right ear only ., All hearing tests were performed in a separate MAC-1 soundproof chamber to eliminate both environmental and electrical noise ., For each HMDP strain , both cochleae from each 8-week-old mouse were removed ., The inner ear was micro-dissected and the surrounding soft tissue and the vestibular labyrinth was removed ., The dissected cochleae were then frozen in liquid nitrogen and then ground to powder ., RNA was extracted and purified by placing cochlea samples in RNA lysis buffer ( Ambion ) ., The sample was incubated overnight ( 4°C ) , centrifuged ( 12 , 000g for 5 minutes ) to pellet insoluble materials and RNA isolated ( following manufacturer’s recommendations ) ., This procedure generates approximately 300 ng of total RNA per mouse ., Illumina’s Mouse whole genome expression , BeadChips , was used for the gene expression measurements ., Amplifications and hybridizations were performed according to Illumina’s protocol ( Southern California Genome Consortium microarray core laboratory at UCLA ) ., RNA was reverse transcribed to cDNA using Ambion cDNA synthesis kit ( AMIL1791 ) and then converted to cRNA and labeled with biotin ., Further , 800ng of biotinylated cRNA product was hybridized to prepare whole genome arrays and was incubated overnight ( 16–20 hrs ) at 55°C ., Arrays were washed and then stained with Cy3 label ., Excess stain was removed by washing and then arrays were scanned on an Illumina BeadScan confocal laser scanner ., EMMA is a statistical test for association mapping correcting for genetic relatedness and population structure and consider the mean per strain and also individual measurement per mouse to increase the statistical power ., We have previously demonstrated that p <0 . 05 genome-wide equivalent for GWA using EMMA in the HMDP is P = 4 . 1×10-6 ( −log10P = 5 . 39 ) 18 ., An R package implementation of EMMA is available online at http://mouse . cs . ucla . edu/emma ., RefSeq genes were downloaded from the UCSC genome browser ( https://genome . ucsc . edu/cgi-bin/hgGateway ) using the NCBI Build37 genome assembly to characterize genes located in each association ., EMMA was used to calculate association ( P-values ) for the probes corresponding to the RefSeq genes ., The confidence interval ( 95% ) for the distribution of distances between the most significant and the true causal SNPs , for simulated associations that explain 5% of the variance in the HMDP , is 2 . 6 Mb 11 ., Only SNPs mapping to each associated region were used in this analysis ., We selected SNPs that were variant in at least one of the HMDP classical inbred strains ., Non-synonymous SNPs within each region were downloaded from the Mouse Phenome Database ( http://phenome . jax . org/ ) ., The generation and initial characterization of Nox3het allele was previously described 19 ., The Nox3het allele arose spontaneously ( endogenous retroviral insertion into intron 12 ) on the GL/Le strain , but has since been made congenic onto the C57BL/6JEiJ strain ., To circumvent the probability of additional alleles from the donor strain this congenic region was backcrossed for more than 10 generations ., Since the downless mutant allele is not present in this strain the congenic interval containing Nox3 is likely less than 5 centimorgans ( http://jaxmice . jax . org/strain/002557 . html ) ., Nox3het ( known as the head-tilt or het mice ) carry autosomal recessive , spontaneous mutations that lead to otoconial absence with no apparent abnormalities in other organs ., The otoconia deficit results in head-tilting behavior and absent vestibular-evoked potentials ( VsEPs ) but normal thresholds ABR 20 ., Pre exposure ABR , DPOAE and VsEP in male and female mice ( 5 weeks old ) of varying Nox3het genotype ( Nox3het/Nox3het and Nox3het/+ ) and wild-type ( C57BL/6JEiJ strain ) was measured as described above ., Pre-exposure threshold levels were obtained at 1 week prior to noise exposure and the animals were assessed for noise damage 2 weeks after exposure ., The ABR permanent threshold shift ( PTS ) was defined as the difference between pre-exposure and post-exposure thresholds at each tested frequency ., One-way ANOVA was used to test the significance and post hoc Tukey test for multiple comparisons ., Mice were sacrificed less than 24 hours after the post exposure ABR ., Cochleae were dissected from the surrounding tissues and openings were made into the coils by piercing the apex and rupturing both the oval and the round windows ., The dissection was done in cold PBS ., After dissection , cochleae were fixed in 4% paraformaldehyde for overnight at 4°C and then washed with PBS ., Further dissection was done to expose the organ of Corti ., For permeabilization and blocking , tissue was immersed for 1 hour in PBS containing 0 . 2% Triton X-100 ( Sigma Chemical ) and 16% normal goat serum ( SouthernBiotech ) ., Samples were incubated overnight at room temperature with primary antibodies ( rabbit anti-myosin6 , 1:500 , Proteus Biosciences and purified mouse anti-CtBP2 , 1:500 , BD Biosciences ) for doubled-staining ., Secondary antibody was then applied and tissue was incubated in the dark overnight ( Alexa 594 donkey anti-rabbit , 1:500 , Life technologies and Alexa Fluor-488 anti-mouse , 1:500 , Life technologies ) ., After , samples were washed three times in PBS and mounted on glass slides using Fluoromount G ( SouthernBiotech ) ., Microscopy was carried out with a laser confocal microscope ( Olympus IX81 ) with epifluorescence light ( Olympus Fluoview FV1000 ) ., Outer hair cell loss ( % per 100μm ) was counted and plotted as cytocochleogram by relating distance of cochlear apex to the tonotopic map of mice of strain CBA 21 ., Percentages indicate the normalized location of the inner and outer hair cells in the cochlea ( 0% , apical and 100% , basal end ) in 10% steps ., Synaptic ribbon density was plotted for each correspondent ABR frequency ( 4 , 8 , 12 , 16 , 24 and 32 kHz ) against the same tonotopic map ., Inner hair cells were analyzed in a row ( 50 μm ) for each frequency ., CtBP2 immunofluorescence spots were counted in z-stacks and divided by the number of inner hair cells ( measured as the quantity of nuclei ) in the sample ., Polymerase chain reaction ( PCR ) was performed for Nox3 using the following primers: Nox3-int12F , GTTCTGGAGCACCACCTTGT; Nox3-int12R CCCATAGGGAGCCAAGAAAT; and ERV-R , TGTCAAGCTGACTCCACCAG 19 ., PCR products were separated on a 1 . 5% agarose gel containing 0 . 5 mg/ml ethidium bromide ., In an effort to identify genomic regions associated with NIHL susceptibility , we phenotyped 5-week old female mice ( n = 297 ) from 64 HMDP strains ( n = 4–5/strain ) for thresholds after noise exposure using Auditory Brainstem Response thresholds at specific ABR stimulus frequencies ., The stimuli consisted of 4 , 8 , 12 , 16 , 24 and 32 kHz tone bursts ., A wide range of ABR thresholds were observed across the HMDP with differences of 3 . 22-fold between the lowest and the highest strains for thresholds at 8 kHz post-noise exposure ( Fig . 1 ) ., Frequencies of 4 , 12 , 16 , 24 and 32 kHz demonstrated differences of 1 . 55 , 3 . 25 , 3 . 57 , 2 . 74 and 3 . 75-fold , respectively ., EMMA algorithm was applied to each phenotype separately to identify genetic associations for the six tone-burst stimuli 18 ., Adjusted association p-values were calculated for 108 , 064 SNPs with minor allele frequency of > 5% ( p < 0 . 05 genome-wide equivalent for GWA using EMMA in the HMDP is p = 4 . 1 x 10-6 , -log10P = 5 . 39 ) ., At this threshold , genome-wide significant associations on Chr ., 2 ( rs27972902; p = 8 . 6x10-7 ) and Chr ., 17 ( rs33652818; p = 2 . 3x10-6 ) were identified for the 8 kHz stimuli ( Table 1 , Fig . 2 ) ., Additionally , a significant association signal on Chr ., 15 ( rs32934144; p = 1 . 7x10-6 ) was identified for the 16 kHz tone burst and two significant regions on Chr ., 3 ( rs30795209; p = 5 . 5x10-7 ) and Chr ., 15 ( rs32278602; p = 5 . 9x10-7 ) were identified at 32 kHz ., Within each association peak there were 4 ( Chr . 15 ) , 11 ( Chr . 3 ) , 10 ( Chr ., 17 ) and 2 ( Chr ., 2 ) unique RefSeq genes ., We next identified genes within each of the five intervals possessing functional alterations ., Genes were selected based upon their regulation by a local expression QTL ( eQTL ) in the HMDP or if they harbored a non-synonymous ( NS ) SNP that was predicted to have functional consequences ., For the eQTL analysis , we generated gene expression microarray profiles using RNA isolated from cochleae in 64 HMDP strains ( n = 3 arrays per strain ) ., EMMA was then used to perform an association analysis between all SNPs and array probes mapping within each region ., A total of 18 , 138 genes were represented by at least one probe , after excluding probes that overlapped SNPs , present among the classical inbred strains used in the HMDP ( see Methods ) ., Of these , 6 genes ( 4 within Chr . 3 association and 2 within Chr . 17 association ) were identified with at least one probe whose expression was regulated by a local eQTL ( Table 2 ) ., However , the only probe whose expression was regulated by a significant local eQTL in the cochlea was located on Chr ., 17 ., We determined whether any of the 27 genes implicated in our preliminary GWAS had a defined role in the inner ear ., The associations on Chr ., 2 , 3 and 15 did not harbor known cochlear genes ., Only NADPH oxidase 3 ( Nox3 ) on Chr ., 17 had been implicated in inner ear biology with mutants lacking otoconia in the utricular and saccular maculae 22 and its high expression in the inner ear 23 ., Of all genes at the chromosome 17 locus , one gene , Tfb1m , had a significant ( 1 . 08x10-6 ) eQTL ( Fig . 3 ) ., Of note , Nox3 , the gene in which our peak GWAS SNP is located , does not have an eQTL in the cochlea; however , there was a clear demonstration 23 that Nox3 is highly expressed ( at least 50-fold higher than in any other tissues ) in specific portions of the inner ear ., Based on these data and the location of our peak GWAS SNP ( rs33652818 ) , we focused on Nox3 as a plausible candidate gene for NIHL at the chromosome 17 locus ., To directly test the hypothesis that Nox3 was associated with susceptibility to NIHL we characterized previously generated Nox3het mice for pre- and post-noise exposure ABR thresholds and PTS after 4 , 8 , 12 , 16 , 24 and 32 kHz tone-burst stimuli ., Consistent with our original GWAS finding , this analysis revealed a statistically significant reduction in the PTS in wild-type mice ( C57BL/6JEiJ strain ) compared to Nox3het/+ and Nox3het/Nox3het at 8 kHz ( Fig . 4 ) ., As a comparison , the effects of the peak SNP ( rs33652818 ) at the Nox3 locus on ABR at various frequencies is shown in Fig . 5 ., Interestingly , there were significant differences as a function of genotype at both the 4 kHz and the 8kHz test frequencies , although the level of significance at 4 kHz ( p = 1 . 1x10-4 ) is only suggestive ( S1 Fig ) and does not reach genome-wide significance ( Table 1 ) ., Thus , the significant and highly suggestive association of rs33652818 with ABR at 8 and 4 kHz , respectively , in the HMDP , as well as the frequency-specific phenotype exhibited by the Nox3het/Nox3het mice , suggests that Nox3 may be involved in NIHL at the lower end of the frequency spectrum ., For a detailed analysis of the entire auditory pathway , we next evaluated outer hair cell ( OHC ) activity using DPOAE and the inner hair cell ( IHC ) and neuronal responses by ABR wave I peak-to-peak amplitudes ., Despite the absence of a statistically significant difference in DPOAE thresholds ( Fig . 6A ) at 8 , 16 , 22 and 32 kHz , there was a pronounced difference at 8 kHz in the wave 1 ABR peak-to-peak amplitudes ( Fig . 6B ) ., The DPOAE ( Fig . 7A ) suprathreshold amplitudes ( dB SPL ) and ABR wave 1 amplitudes ( μV ) ( Fig . 7B ) for the 8 kHz tone burst were compared at different stimulus intensities ., Both analyses demonstrated statistically significantly less noise damage in the wild-type in comparison to the heterozygous and mutant mice ., To confirm these electrophysiological findings , we collected cochleae from pre- and post-noise exposure Nox3het mice and wild-type ., First , we assessed OHC loss throughout the entire cochlea by creating a cytocochleogram ( Fig . 8A ) of immunolabeled ( Fig . 8B ) whole-mount organs of Corti to correlate with the DPOAE findings ., Subsequently , the IHC afferent synaptic density ( Fig . 9 ) was analyzed as a marker of the neuronal responses ( suprathreshold ABR wave 1 amplitude ) ., Despite the absence of a statistical significance in OHC loss , the Nox3het/+ and Nox3het/Nox3het mice demonstrated a significantly reduced post-noise exposure density of synaptic ribbons ( at the 8kHz tonotopic location ) ., We have , for the first time , used association analysis with correction for population structure to map several loci for hearing traits in inbred strains of mice ., Our results identify a number of novel loci for susceptibility to NIHL ., Additionally , our study demonstrates frequency-specific genetic susceptibilities to noise within the cochlea and the power of our GWAS to detect frequency-specific loci that are precisely recapitulated in a mutant mouse model ., Mouse GWAS has revolutionized the field of genetics and has lead to the discovery of hundreds of genes that are involved in complex traits 24 ., Our successful mapping largely came from the initial observation that there was a clear strain variation at all post noise exposure hearing phenotypes , reiterating the contribution of genetic factors to NIHL susceptibility ., This wide distribution of phenotypes and genotypes facilitated our high-resolution genetic mapping ., We used a combined set of 64 classic inbred and recombinant inbred strains , a portion of the HMDP , as an extension of the classical inbred strain association ., This increased the statistical power of the classical association studies by including a set of recombinant inbred strains in the mapping panel 25 ., The HMDP provided significant statistical power and resolution to identify a locus for NIHL susceptibility that was precisely modeled in a mutant strain 26 ., Although this panel is composed of 100 commercially available inbred strains , with roughly two-thirds of this panel we were able to map 5 loci , reflecting the power to detect loci with moderate effect ., In addition to the power present in this resource , the resolution of this panel is , in some cases , two orders of magnitude better than that achieved with linkage analysis , as we have recently demonstrated in our mouse GWAS for age-related hearing loss 27 ., In an unprecedented manner , this new paradigm was applied to the first high-resolution mapping of candidate genes for NIHL susceptibility ., Our GWAS generated significant associations in at least five loci at three different post-noise exposure stimulus frequencies , corresponding to a total 27 candidate genes ., All of these candidate genes require adequate characterization , but the first gene to be validated by a genetic mutant mouse model was Nox3 ., Nox3 was selected for further investigation based upon its relatively restricted expression in the cochleo-vestibular epithelium and spiral ganglion neurons 23 ., The Nox3 gene was described in 2000 based upon its sequence similarity to other Nox isoforms ( encodes an NADPH oxidase ) 28 ., The overall structure of Nox3 is highly similar to that of Nox1 and Nox2 29 and Nox3 shares 56% amino acid with Nox2 30 ., Encoded by Nox3 , the six-transmembrane NADPH-binding protein interacts with a two-transmembrane protein ( encoded by Cyba ) and a cytosolic protein ( encoded by Noxo1 ) ., This activation releases a functional NADPH oxidase complex that is able to transporting electrons across membranes towards oxygen ( O2 ) generating superoxide ( O2•- ) and subsequent reactive oxygen species ( ROS ) 19 ., First studies on the Nox3 function were published in 2004 and generated the definition of Nox3 as an NADPH oxidase of the inner ear 2322 ., Banfi , et al . , performed analysis of Nox3 distribution ( real time PCR and in situ hybridization ) and reported high Nox3 expression in the inner ear ( cochlear/vestibular sensory epithelia and the spiral ganglion ) ., Following exposure to cisplatin , HEK293 cells transfected with Nox3 produced O2•- spontaneously and generated a dramatic increase in O2•- production 23 ., Paffenholz et al . 22 reported that mutations of the het locus affect Nox3 and that these head tilt mice ( het ) have impaired otoconial formation in the utricle and saccule resulting in balance defects , such as the inability to detect linear acceleration or gravity ., Based upon this finding we chose to pursue interrogation of Nox3 , a gene within our locus on Chr ., 17 ., Subsequent studies have established a role for the Nox3 gene as the primary source of ROS generation in the cochlea , especially induced by cisplatin ototoxicity 31 ., The knockdown of Nox3 ( pretreatment with siRNA ) prevented cisplatin ototoxicity with preservation of hearing thresholds and hair cells ., Also , it reduced the expression of Nox3 and biomarkers of damage ( TRPV1 and KIM-1 ) in cochlear tissues 32 ., siRNA-mediated gene silencing of Nox3 alleviated cisplatin-induced hearing loss in rats and reduced apoptosis of the sensory hair cells in the cochlea 33 ., Although there was no similar evidence regarding NIHL , this key role for Nox3 in the development of cisplatin ototoxicity confirming its role in regulatory mechanisms of cochlear damage encouraged us to validate this candidate gene for NIHL ., The only study exploring NIHL and the NOX family ( including Nox3 ) was completed in rats 34 ., This study did not indicate whether the Nox3 gene decreased or increased the susceptibility to noise , but instead it evaluated Nox3 expression levels after noise exposure ., Some members of the NADPH oxidase family ( Nox1 and Duox2 ) were up-regulated in the rat cochlea after noise exposure , suggesting that these isoforms could be linked to cochlear injury ., In contrast , the Nox3 isoform was down-regulated after exposure to 100 dB SPL and 110 dB SPL by seven and fivefold respectively , which could represent an endogenous protective mechanism against oxidative stress ., This protective mechanism may have decreased the impact of the noise among wild-type rats by reducing the expression of Nox3 and decreasing the difference related to mutants ., However , the in vivo data was based on the use of a non-specific Nox inhibitor that targeted multiple members of this enzyme without conclusively demonstrating that Nox3 plays a role in NIHL ., Our study , by contrast , has used animal models with naturally occurring genetic variation and specific genetic perturbation of Nox3 to directly implicate this oxidative stress enzyme in hearing ., According to our study , noise exposure might have an opposite effect to cisplatin on Nox3 expression , suggesting differential involvement of Nox3 on noise and cisplatin-induced cochlear damage ., Based upon this literature we hypothesized that the absence or reduction of the Nox3 gene product , responsible for the production of ROS in the cochlea , would reduce susceptibility to noise and were startled by our findings ., A review of the literature shows there are several key protective mechanisms attributed to the Nox family of genes ., These mechanisms include: host defense and inflammation ( ROS-dependent killing , inactivation of microbial virulence factors , regulation of pH and ion concentration in the phagosome and anti-inflammatory activity ) , regulation of gene expression ( TNF-alpha , TGF-beta1 and angiotensin II ) , cellular redox potential , cellular signaling ( inhibition of phosphatases , activation of kinases , regulation of ion channels and Ca2+ signaling ) , oxygen sensing ( kidney , carotid body and lungs ) , biosynthesis , regulation of blood pressure , cell growth , angiogenesis , differentiation and senescence 30 ., These protective mechanisms may very well play a role in the findings of susceptibility to NIHL in the wild-type animals ., We were able to validate our frequency-specific GWAS findings in isolation by studying Nox3het mutant mice ., After noise exposure there was a statistically significant difference between the wild-type mice in comparison to the homozygous mutants and the heterozygotes on several measures of auditory function specifically and solely after the 8 kHz exposure ., Contrary to the initial expectations , the presence of the Nox3 gene was clearly protective against noise damage ., Also we were able to demonstrate the genotypic effect of the peak SNP at the same GWAS phenotype at 8 kHz ., We also show genotypic effect on 4 kHz , but this finding was only suggestive in GWAS and not confirmed in Nox3het mutants ., We dissected this phenotype in detail physiologically by assessing OHC function using DPOAEs and IHC/auditory nerve function using ABR ., Although there was no statistically significant difference in DPOAE thresholds amongst the genotypes , there was a marked difference in the amplitude of wave 1 of the ABR after suprathreshold stimulation with the 8 kHz tone burst ., This suggested that the mechanism of hearing loss , in relation to Nox3 , resided in the spiral ganglion neurons and likely at 8 kHz along the cochlear place map ., There are many genes differentially expressed along the tonotopic axis of the cochlea , and this has been shown for Nox3 35 ., It is likely that our frequency specific finding of variation in susceptibility to NIHL is the result of this tonotopic expression pattern ., Considering that all of the results pointed to the area of 8 kHz , we initiated a thorough electrophysiological and histological dissection at this particular frequency ., The evaluation of the DPOAEs and suprathreshold wave 1 ABR amplitudes was performed at multiple stimulus intensity levels ., For each study , the wild-type were more resistant to NIHL at only at 8 kHz ., We performed immunohistochemistry two weeks after the noise exposure ., Although the difference in OHC loss was not significant , we demonstrated a significantly higher density of synaptic ribbons in wild-type mice ., Thus , the electrophysiological findings were verified by the immunohistochemistry , demonstrating that the presence of Nox3 is protective at the neuronal level and that the sensory neural hearing loss after noise exposure occurred at this level of the peripheral auditory system ., The absence of differences in outer hair cell count was also verified by its corresponding electrophysiological measure of DPOAE thresholds ., However , through the evaluation of DPOAE suprathreshold amplitudes , we were able to observe a statistically significant higher amplitude in the wild-type mice ., These three different measures of the integrity of the outer hair cells ( outer hair cell count , DPOAE thresholds and DPOAE suprathresholds amplitudes ) have different sensitivity profiles to demonstrate the impact of noise ., Probably DPOAE suprathreshold amplitude is the most sensitive measurement , since there is greater signal-noise ratio ., This metric indicates that there is significantly less impact on the activity of the outer hair cells in wild-type mice ., Although Nox3 is associated with production of O2•- in the inner ear , the Nox family has several physiological and potentially protective mechanisms ., Definitely , this protective role explains the fact that the absence of

Introduction, Methods, Results, Discussion

In the United States , roughly 10% of the population is exposed daily to hazardous levels of noise in the workplace ., Twin studies estimate heritability for noise-induced hearing loss ( NIHL ) of approximately 36% , and strain specific variation in sensitivity has been demonstrated in mice ., Based upon the difficulties inherent to the study of NIHL in humans , we have turned to the study of this complex trait in mice ., We exposed 5 week-old mice from the Hybrid Mouse Diversity Panel ( HMDP ) to a 10 kHz octave band noise at 108 dB for 2 hours and assessed the permanent threshold shift 2 weeks post exposure using frequency specific stimuli ., These data were then used in a genome-wide association study ( GWAS ) using the Efficient Mixed Model Analysis ( EMMA ) to control for population structure ., In this manuscript we describe our GWAS , with an emphasis on a significant peak for susceptibility to NIHL on chromosome 17 within a haplotype block containing NADPH oxidase-3 ( Nox3 ) ., Our peak was detected after an 8 kHz tone burst stimulus ., Nox3 mutants and heterozygotes were then tested to validate our GWAS ., The mutants and heterozygotes demonstrated a greater susceptibility to NIHL specifically at 8 kHz both on measures of distortion product otoacoustic emissions ( DPOAE ) and on auditory brainstem response ( ABR ) ., We demonstrate that this sensitivity resides within the synaptic ribbons of the cochlea in the mutant animals specifically at 8 kHz ., Our work is the first GWAS for NIHL in mice and elucidates the power of our approach to identify tonotopic genetic susceptibility to NIHL .

Noise-induced hearing loss ( NIHL ) is the most common work-related disease in the world and the second cause of hearing loss ., Although several candidate gene association studies for NIHL in humans have been conducted , each are underpowered , un-replicated , and account for only a fraction of the genetic risk ., Buoyed by the prospects and successes of human association studies , several groups have proposed mouse genome-wide association studies ., The environment can be carefully controlled , facilitating the study of complex traits like NIHL ., In this manuscript , we describe , for the first time , an association analysis with correction for population structure for the mapping of several loci for susceptibility to NIHL in inbred strains of mice ., We identify Nox3 as the associated gene for susceptibility to NIHL that the genetic susceptibility is frequency specific and that it occurs at the level of the cochlear synaptic ribbon .

journal.pcbi.0030041

UV-Induced Mutagenesis in Escherichia coli SOS Response: A Quantitative Model

The SOS response in the bacterium E . coli encompasses many proteins involved in detecting and repairing DNA damaged by a variety of agents , such as UV radiation , or chemicals such as mitomycin and bleomycin 1 ., A complex regulatory network , comprising both transcriptional and post-translational regulators , controls the concentrations and levels of activity of these proteins ( Figure 1 . ) The collective actions of this regulatory network are orchestrated so that the SOS response is commensurate with the magnitude of DNA damage 1 ., Mutagenesis , such as the introduction of single-base substitutions in the DNA sequence , is not an inevitable consequence of DNA damage , but results from the action of specialized error-prone DNA polymerases that are part of the response 2 ., This constitutes an extreme measure that might be useful for the cell only after very heavy DNA damage when DNA replication and repair cannot effectively proceed without it ., While some mutations might benefit the offspring , the vast majority is harmful; therefore , the presence of error-prone polymerases should be tightly regulated to prevent their action at low doses of UV ., Briefly , the sequence of events triggered by UV irradiation of E . coli is as follows: UV radiation damages the DNA by creating lesions that mechanically disrupt the process of DNA duplication by stalling the DNA-polymerase ( Pol III ) in a moving replication fork ., This , in turn , results in the production of single-stranded DNA ( ssDNA ) gaps ., These gaps are coated by the protein RecA 1 , 3 , 4 , forming long nucleoprotein filaments in which it assumes its active form , RecA* ., RecA* , together with other proteins , is involved in the nonmutagenic filling in of ssDNA gaps via homologous recombination 5 , and it catalyses the cleavage of the transcriptional repressor LexA 6 and of the protein UmuD 7 , whose cleaved form—UmuD′—is necessary for mutagenesis 1 ., The drop in the level of the transcription factor LexA , due to its cleavage , de-represses the regulon involved in the SOS response ., This regulon comprises about 30 genes , including those encoding the mutagenesis proteins UmuD and UmuC , RecA , and LexA itself ., Also part of the SOS regulon are genes encoding UvrA , B , C—a group of nucleotide excision repair ( NER ) proteins that locate and excise damaged regions from the DNA 8 , 9 ., Mutagenesis in UV-irradiated E . coli cells is mainly the direct result of the activity of the error-prone DNA polymerase , Pol V 2 ., Pol V consists of two units of UmuD′ and one unit of UmuC ., It inserts several random base pairs in the DNA strand directly opposite a lesion , thus helping a replication fork to quickly bypass the lesion , after which Pol III can take over and continue replication ., A distinct coordinated subnetwork of proteins centered on UmuD and UmuC controls the abundance , and thereby the activity , of Pol V ( Figure 1 ) ., Even though the SOS response in bacteria has been studied for several decades , new discoveries continue to be made ., Recent single-cell experiments measured the temporal dependence of the activity of LexA-regulated promoters 10 , which showed the following features: For low UV doses , the promoter activity peaks at about 10 min after the UV dose ., This was also observed in bulk measurements of promoter activity averaged over a large population of cells 4 and can be attributed to the initial rapid drop in LexA levels after UV damage because of the activation of RecA , followed by a slow increase to its original level as the lesions are repaired by NER and the level of RecA* falls ., More surprising was the observation that at higher doses of radiation , LexA-regulated promoter activity often had a second peak at about 30–40 min , sometimes even followed by a third peak at 60–90 min ., This resurgence of the SOS response is puzzling because it indicates a temporary increase in RecA* levels at a time when the NER process is well under way and the number of lesions are already falling ., This second peak ( but not the third peak ) was , however , absent in both ΔUmuDC null-mutants and mutants that have an uncleavable version of UmuD ( K97A ) 10 ., The common element in both types of mutants is the absence of Pol V , which suggests that the second peak is related to mutagenesis ., In this paper we propose a plausible mechanism for the appearance of this peak ., We argue that E . coli bacteria can reliably measure the total amount of DNA damage ., The ability of replication forks to bypass bulky lesions allows the cells to “count” the number of lesions they encountered over a fixed time interval ( the average lifetime of RecA* filaments ) ., The result of this count , given by the instantaneous number of RecA* filaments , is then fed into the mutagenesis regulatory subnetwork , which—as we show below—is designed to time-integrate this input signal over a long interval ( 30–40 min ) and to abruptly turn on the Pol V if the integrated level of damage exceeds some critical threshold ., The appearance of Pol V speeds up the bypass of lesions , and thus increases the rate at which new lesions are encountered by replication forks ., We believe that this positive feedback from Pol V to the RecA* concentration is responsible for a temporary increase in the activity of SOS-regulated promoters 30–40 min after the radiation ( the second peak reported in 10 . ), The goal of this paper is to model temporal dynamics of the mutagenesis subnetwork of the SOS response system ( highlighted in yellow in Figure 1 ) for different doses and durations of UV radiation ., This subnetwork is not isolated from the rest of SOS response , and therefore the model includes other parts of the entire E . coli regulatory network that interact with proteins involved in mutagenesis ., Figure 1 shows the components of the SOS response that we quantify in our model ., Different colored arrows correspond to different mechanisms of interactions between the nodes ., An excellent earlier paper by Aksenov 11 contains a model of LexA-controlled transcriptional regulation coupled with the NER repair of lesions during the SOS response ., Here that model is extended to incorporate the mutagenesis subnetwork ., Full details of our model and parameter values are provided in the Methods section ., We mathematically model the temporal dynamics of the density of UV-induced lesions , as well as concentrations of LexA , RecA* , unbound UmuD , unbound UmuD′ , UmuD–UmuD′ heterodimer , and Pol V , using a set of ordinary differential equations ., Positive and negative terms in these equations represent different ways of production and consumption/degradation of the corresponding quantities ., We do not explicitly simulate the creation and repair of individual lesions , nor do we simulate each replication fork moving along the DNA ., Thus , our model ignores stochastic fluctuations ., However , in later sections we do examine the effect of averaging over a population of cells in which various parameters , e . g . , the number of replication forks , vary from cell to cell ., This provides an in silico comparison between single-cell and cell-culture measurements ., We also treat all time delays , such as when a replication fork is stalled at a lesion , in a simplified manner , i . e . , we assume that these delays affect the RecA* level only via the average replication speed ., Most parameters in our model have been fixed using experimental data ., For example , the experiments in 3 , 4 , 12 allow us to fix the RecA*-mediated cleavage rates of LexA and UmuD ., The model has a total of 18 parameters of which only three could not be fixed by experimental data ., We have therefore scanned a range of reasonable values for these three , as described in a later section ., Our model indicates four key features of the mutagenesis subnetwork in E . coli: 1 . A mechanism for measuring the local amount of damage , coupling the number of RecA* filaments to the current lesion density ., 2 . A long-term “memory” used to time-integrate the RecA* signal and thus to determine whether the damage level remained high for a substantial time ., This mechanism is based on slow accumulation of UmuD′ ., 3 . Strong binding between UmuD and UmuD′ , which provides a highly ultrasensitive increase in unbound UmuD′ levels as its concentration exceeds that of its “inhibitor” UmuD ., 4 . Positive feedback from Pol V to RecA* levels , which further increases the sharpness of the turn-on and turn-off of Pol V . This mechanism is also responsible for the second peak in activity of SOS promoters ., In the subsequent sections , we discuss each of the above aspects in more detail ., First we propose the following mechanism for the influence of the UV dose on the RecA* level ., Consider a given replication fork proceeding on a DNA strand that has UV-induced lesions , as depicted in Figure 2 . The Pol III DNA-polymerase stalls at the first lesion , generating an ssDNA gap that is then covered with RecA ., This RecA filament exists for an average time , denoted τRecA* , after which it disassembles ., ( We assume that each filament disassembles independently with a rate that is not limited by other DNA damage–induced processes . ), During this time the replication fork may bypass the lesion and continue processing the DNA , leaving the first RecA filament behind ., If the time the fork spends stalled at a lesion is sufficiently large or the lesion density is sufficiently small ( so that the time the fork spends traveling between lesions is large ) , then the first filament will disassemble before the fork reaches the next lesion and creates another filament ( as in Figure 2A ) ., Therefore , in this case , there will be no more than one RecA* filament per replication fork at any time ., On the other hand , if the stall time is small or the lesion density is large , the fork will reach a second lesion before the first filament disassembles and , as a consequence , there may be many RecA* filaments per fork existing simultaneously on the DNA ( as in Figure 2B ) ., The RecA* level directly depends on the time a polymerase spends traveling between lesions , τmoving = 1/μν , where μ is the density of lesions on the chromosome , and v is the average speed with which Pol III processes DNA replication on undamaged DNA ., This dependence can be quantified: one RecA* filament is produced every time the replication fork encounters a lesion ., If the fork spends time τstalled at a lesion and time τmoving between lesions , then the rate of production of RecA filaments is given by the following formula:, Further , the filament disassembly rate is, , where Nfil is the number of RecA* filaments associated with the replication fork under consideration and τRecA* is the average persistence time of a RecA* filament ., Because the rates of filament production and disassembly are much faster than all other processes we are interested in ( the transcription of SOS genes and the rate of NER repair ) 13 , we can assume that the number of RecA* filaments at any given time are such that the production rate equals the disassembly rate , i . e . ,, The total amount of RecA* , r* , is given by the above expression multiplied by LRecA*—the average length of a RecA* filament ( taking into account the finite probability of forming a filament at each lesion a fork encounters ) —and Nf , the total number of replication forks currently duplicating DNA in a cell , i . e . ,, After fixing the parameter values based on experimental data ( see Methods ) , this relation gives a RecA* level of approximately 100 nM for a fixed lesion density produced by a UV dose of 2 J/m2 , while it gives more than 400 nM for a UV dose of 50 J/m2 ( this neglects the effects of Pol V , which will be discussed later ) ., The process shown in Figure 2 is thus a simple way for the cell to “count” the number of lesions on the DNA using a “memory , ” which is the finite existence time of a RecA filament ., This is a short-time memory lasting only for a time τRecA* ., However , the rate of UmuD′ production is proportional to the amount of RecA* , therefore the UmuD′ level is a measure of RecA* level integrated over time ., Thus , UmuD′ accumulates if damage ( and therefore RecA* ) persists for a long time ., In our model , with RecA* at its maximum possible level , the timescale for the UmuD′ level to exceed that of UmuD is about 15 min ., For smaller UV doses , and therefore lower RecA* , this rise time can be more than 35 min ., UmuD′ is an integral component of the error-prone polymerase Pol V . However , UmuD′ has to accumulate to a fairly high level before Pol V appears in any detectable quantities ., The main reason for this is a strong physical interaction between UmuD and UmuD′ ., The binding between them is stronger than that between UmuD or UmuD′ pairs; when UmuD and UmuD′ are mixed in equimolar concentrations , the heterodimer is found to be much more abundant than either homodimer ( UmuD–UmuD and UmuD′–UmuD′ ) 14 ., This strong binding ensures that unbound UmuD′ homodimers required for Pol V formation appear in sufficient quantities only when ( and if ) the total concentration of UmuD′ exceeds that of UmuD ., Figure 3 shows the equations we use to model the dynamics of UmuD , UmuD′ , and Pol V . These equations model the following processes: ( 1 ) LexA represses the production of UmuD ( βu , Ku ) ; here , we assume a Hill coefficient of 1 based on the fact that the upstream region of the UmuD promoter has only one LexA binding site 15; ( 2 ) RecA* catalyzes the intermolecular cleavage of UmuD 16 ( of both free and heterodimer forms ) to produce UmuD′ at rate γu; ( 3 ) UmuD and UmuD′ form a heterodimer 14 with on- and off-constants given by Kf , and Kb; ( 4 ) ClpX degrades UmuD′ ( but not UmuD ) when it is in the heterodimer 17 , at rate γdd′; ( 5 ) All molecules are diluted by cell growth and division ( γdil ) ., Pol V is composed of two units of UmuD′ bound with one unit of UmuC protein ., Thus , the level of Pol V cannot exceed that of UmuC ( C ) , but for small amounts of UmuD′ it is proportional to u′2 ., K controls how much of the UmuD′ homodimer is required to saturate the levels of Pol V . The UmuC concentration C for simplicity is assumed to be constant during the narrow time window where it matters ( i . e . , when u′ is nonzero ) ., The qualitative aspects of the dynamics produced can be understood by looking at a simplified version of these equations: since RecA* levels change relatively slowly , first consider UmuD and UmuD′ levels at a fixed RecA* concentration , and thus a constant UmuD → UmuD′ cleavage rate γur* ., If the heterodimerization is extremely strong , the time course of the total ( free + heterodimer ) UmuD′ (, = u′ + uhetd ) satisfies the following rate equation ( see Methods for the derivation from the equations in Figure 3 ) :, Here utot = u + uhetd is the total concentration of noncleaved UmuD ( in free or heterodimer form ) ., The first term , γur*utot , is the production of UmuD′ due to the cleavage of UmuD , the second term is the ClpX-dependent degradation of UmuD′ inside UmuD′–UmuD heterodimers , while the last term is the decrease in the concentration of UmuD′ due to cell growth and division ( the dilution term ) common for all proteins in the cell ., With LexA and RecA* levels fixed , i . e . , γur* constant , we can calculate the steady-state levels of UmuD and UmuD′ from these equations and , hence , the condition for Pol V to be present , i . e . , when UmuD′ exceeds UmuD:, > utot ., Setting, = 0 and, = utot , we obtain the condition for, > utot in the steady state:, independent of UmuD production and degradation rates ., Thus , Pol V abruptly appears once the RecA* level , and hence the value of γur* , crosses and stays above the required threshold for long enough to allow UmuD′ to accumulate and pass the UmuD level ., This analysis also suggests that there would be a threshold minimum UV dose below which Pol V does not appear because the NER repair brings down DNA damage quickly enough to bring the level of RecA* below the amount required to satisfy Equation 3 . The behavior of replication forks at lesions ( described above ) naturally provides a positive feedback from Pol V to RecA* because Pol V reduces the stall time at the lesion , τstalled ( 2 estimates that Pol V bypasses lesions with 100- to 150-fold higher efficiency than Pol III ) ., This is illustrated in Figure 4 . Initially , there is no Pol V; however , other “nonmutagenic” translesion synthesis polymerases , Pol IV and Pol II ( DinB or PolB ) , which are always present in the cell , ensure that even in the absence of Pol V the stalled replication fork could still bypass a lesion 18 at a rate we denote, ., In Figure 4 , this rate is slow enough that by the time the fork reaches the next lesion ( after a time, + τmoving > τRecA* ) , the first filament disassembles ., At a later time , when Pol V appears , the stall time reduces dramatically 2 , 19 ., The scenario depicted in Figure 4 assumes the bypass rate is dominated by Pol V–assisted bypass ( for the more general treatment used in our model , see the Methods section ) ., In this case , the reduction in stall time from, to, when Pol V appears is sufficient to allow the replication fork to reach a second lesion before the first RecA* filament disassembles ., Therefore , the RecA* level rises when Pol V appears ., When this rise is fast enough , which occurs for a large enough UV dose , this results in a second peak in LexA-controlled promoter activities , as shown in Figure 5 . Thus , the second peak is a natural consequence of the mechanism for setting RecA* levels represented by Equation 1 . This prediction of the model is confirmed by the recent single-cell fluorescence experiments of Friedman et al . 10 ., They also found that the second peak was washed out when the signal was averaged over many cells , probably because of cell-to-cell variations ., Among the parameters , which can vary between cells , is the number of replication forks ., We find that averaging the LexA-controlled promoter activity predicted by our model over many cells with differing numbers of replication forks produces a curve with a single peak ( Figure 5 , red dashed line ) as observed in the experiments ., The model reveals an almost digital response of Pol V levels to UV , which provides very tight control of mutagenesis ., Figure 6 shows the predictions of our model for the time course of Pol V ( UmuD′2C ) for different UV doses ., In these simulations , the cell is subjected to an instantaneous pulse of UV at the specified dose at time zero ., The main features of this plot are: ( 1 ) the existence of a UV dose ( about 17 J/m2 ) below which the Pol V level is very low ., Thus , with low damage , mutagenesis is virtually absent and DNA repair is error-free; ( 2 ) a sharp onset in the generation of Pol V at about 15–35 min for UV doses larger than 17 J/m2 ., The time of onset is largely UV-independent at high doses; ( 3 ) a rapid turn-off of Pol V at variable times that increase with the UV dose ., This plot confirms several points suggested by the analysis of the model in the previous sections ., First , the existence of a minimum threshold UV dose below which no Pol V is produced is a consequence of the equations described in Figure 3 and , in particular , Equation 3 . The rapid onset and the later rapid decrease of Pol V is due to the combination of heterodimerization and the previously described positive feedback from Pol V to RecA* levels ., We provide more evidence to support this conclusion in the next section ., The above analysis uses a simplifying assumption that the binding between UmuD and UmuD′ is infinitely strong , so that the level of UmuD–UmuD′ heterodimer is simply given by min ( UmuD′ , UmuD ) ., The model can be used to examine the importance of the strength of this interaction in the mutagenesis response ., Figure 7A illustrates the effect of decreasing this dissociation constant ( Kdd′ ) ., It shows that a strong association is critical in setting the abruptness and positions of both the turn-on and turn-off points for Pol V . Another relevant protein–protein interaction is the binding between UmuC and UmuD′ homodimers to form Pol V ( K ) ., This is one of the parameters for which experimental data are not available ( see Methods ) ., However , Figure 7B shows that decreasing this dissociation constant makes the Pol V profile more “digital , ” i . e . , more step-like with the concentration being either zero or maximum most of the time ., Decreasing the ClpX-dependent degradation rate of UmuD′ in the heterodimer , γdd′ , mostly delays the turn-off of Pol V without affecting its turn-on time ( Figure 7C ) ., Figure 7D shows the effect of turning off the positive feedback from Pol V to RecA* ., Clearly , this feedback , combined with strong heterodimerization , is a crucial ingredient in the rapid onset of Pol V . Without feedback , the Pol V level is an order of magnitude lower compared with when there is feedback ., Another direct implication of Equation 1 is that the peak amount of RecA* saturates as the UV dose is increased ., Indeed , as the density of lesions μ rises , τmoving = 1/μv decreases ., According to Equation 1 , the RecA* level saturates once τmoving becomes much smaller than τstalled ., Consequently , the height of the first peak of LexA-controlled promoter activity eventually saturates at high UV doses ., During the second peak of promoter activity , the RecA* concentration rises again as τstalled drops due to the Pol V–assisted bypass of lesions ., The height of the second peak also saturates , but at higher UV doses ., For the parameters used in our model , the amplitude of the first peak of promoter activity reaches 90% saturation around 25 J/m2 , while that of the second peak reaches it around 48 J/m2 ( see Figure 8B . ) This prediction of our model is in agreement with the experimental data in Figure 4C of 10 , which show that the saturation of the second peak occurs at a higher UV dose than for the first peak ., However , that data show the peak height averaged over a cell population ., Therefore , to compare our model directly with the data , we show in Figure 8A the peak heights averaged more than 200 runs with varying Nf ., The resultant peak height versus UV dose curves match the data of 10 satisfactorily with the exception of the first peak data point at 50 J/m2 , which is lower than the previous data points ., One explanation could be the ambiguity in the averaging procedure because , especially at higher UV doses , the second peak may sometimes be large enough to outswamp the first one , and hence be counted as a first peak , raising the red curve ., Note , however , that at the single-cell level our model will always show a monotonically increasing peak height as UV dose is increased ., The behavior of our model also agrees with Figure 4A of 10 from which we conclude that the second peak of promoter activity starts to appear at a considerable frequency for UV doses between 10 and 20 J/m2 ., The threshold of 17 J/m2 predicted by our model ( the same as the threshold for mutagenesis ) is consistent with this ., The SOS response of bacteria to radiation is typically studied by exposing them to a very short burst of UV light and then following the repair of the DNA damage ., However , in environments for which bacteria are evolutionarily adapted , there may be both short bursts of the UV radiation , similar to the experimental conditions imposed on them , as well as much longer spells of low intensity UV exposure ., The latter type of perturbation might not be well-suited for in vivo experiments but is easily achievable in our in silico model ., Adding a new term representing a continuous low rate of production of lesions ( see Methods ) gives rise to a stable steady state wherein the rate of NER repair equals the rate of creation of the new DNA damage ., Figure 9A shows the typical response to a continuous UV dose , which is low enough that mutagenesis is never triggered; LexA and RecA* take about 60 min to reach a steady state ., Experiments in which cells were exposed to continuous UV damage because of the presence of a constant amount of mitomycin C also indicate that the SOS response ( rates of LexA repressor synthesis and cleavage ) took 60 min to reach a steady state 3 , confirming this prediction of our model ., We also simulated the response of our virtual cell to a pulse of UV radiation of a given integral intensity and duration that varied from 0 min to 300 min ., Figure 9B separates the mutagenic and nonmutagenic regions of parameter space ., Initially , the magnitude of the SOS response weakly increases with prolonging the duration of the pulse ., This response was expected since very short pulses give the NER subsystem time to repair some lesions before replication forks encounter them; therefore , the average RecA* concentration is less than that for slightly longer pulses ., The threshold for activating the mutagenesis subsystem reaches its minimum value for an ∼60-min pulse , and then it increases linearly throughout the duration of the pulse , indicating that the cell has reached the steady state in which mutagenesis is not triggered by the total intensity of the pulse , but rather by a sufficiently high rate of production of new lesions corresponding to a UV intensity per unit time of about 1 . 5 mW/m2 ., This is an order of magnitude less than the typical solar UV intensity of 7–10 mW/m2 in Copenhagen at noon on a clear day in December ., For comparison , the solar UV intensity in the tropics in similar conditions is more than 100 mW/m2 ( see http://www . temis . nl/uvradiation/UVindex . html ) ., When bacteria experience a large amount of DNA damage , their response has a mutagenic component that , it has been suggested , might afford some evolutionary advantage by altering the genome of offspring that would allow some of them to better survive high levels of the damage-inducing agents 20 ., Precursors to an error-prone polymerase have also been implicated in slowing down DNA replication 21 , thereby allowing additional time for accurate repair processes to remove lesions from the DNA ., This delay is immediately terminated once the error-prone polymerases are fully formed ., However , this kind of evolutionary strategy would be harmful where there was no damage , or when it was sufficiently low that it could be quickly repaired by error-free mechanisms ., Hence , mutagenesis must be tightly regulated ., The main features of the mutagenic component of the SOS response system , according to published literature , are the following: ( 1 ) Mutagenesis is characterized by a sharp temporal onset and turn-off and threshold-like behavior as a function of UV dose ., There is strong experimental evidence for this ., For example , Rangarajan et al . 18 observed that in the absence of Pol II masking the effects , Pol V–assisted bypass rapidly appears about 45 min after the irradiation ., Also , from Figure 4 of 21 we may conclude that the UmuD′ concentration becomes comparable to that of UmuD about 30 min after irradiation , irrespective of UV dose ., This exactly matches the time at which Pol V appears in our model when UmuD–UmuD′ binding is very strong ., ( 2 ) Mutagenesis gives rise to the second peak in activity of the SOS regulon ., This is inferred from data in 10 that show this second peak is absent in mutants that lack UmuD or contain an uncleavable version of it ., We constructed a network model of mutagenesis in the bacterial SOS response system to account for these features ., Figure 10 summarizes the key aspects of the behavior of the system that emerged in our simulations ., We demonstrated that strong binding between UmuD and UmuD′ is necessary for the sharp onset of mutagenesis and for its turn-off when UmuD′ again falls below UmuD ( see Figure 7A ) ., Thus , initially , when levels of UmuD′ are low , almost all of the UmuD′ is sequestered in heterodimers so that no Pol V is generated ., However , UmuD′ is being constantly produced by the cleavage of UmuD , whose production , in turn , is elevated due to the de-repression of its promoter ., If the UV damage is large enough , eventually the concentration of UmuD′ rises sufficiently to exceed that of UmuD and allow the formation of Pol V . Additional control is afforded by the degradation of the UmuD–UmuD′ heterodimer by ClpX , which removes UmuD′ while freeing UmuD for further cleavage or dimerization ., Although this degradation is not essential for the systems qualitative behavior , it substantially influences the turn-off time and rate ( Figure 7B ) ., Indeed , without it , turn-off could be only realized by the reduction in UmuD cleavage rates due to DNA repair and would depend solely on the slower NER mechanism ., In addition , Lon actively degrades UmuD homodimers and UmuC 17; its physiological advantages are unclear ., Including this mechanism in our model does not affect the systems qualitative behavior , provided the degradation rate is not too large ., We suggested a simple mechanism by which the RecA* level can serve as a measure of the lesion density ( see Equation 1 ) ., This mechanism relies on the possibility for RecA filaments to exist for some finite time after the replication fork has bypassed the lesion where the filament was created ( note that we assume that this happens whether the lesion was on the leading or lagging strand ) ., This allows the replication fork to sample a stretch of DNA , thus counting the damage density that is then manifested in the RecA* level ., A direct implication of this mechanism is that there is a positive feedback from the Pol V to RecA* levels ( see Figure 4 ) ., The resulting temporary increase in RecA* levels due to the sudden appearance of Pol V is sufficient to explain the resurgence of the SOS response 30–40 min after irradiation , observed in the single-cell experiments of Friedman et al . 10 ., In addition , this mechanism also explains their observation of saturation of the peak promoter activities , and hence RecA* levels , upon increasing the UV dose ( see Figure 8 ) ., Note that the first peak in promoter activity is produced due to changes in the lesion density , and thereby τmoving , as NER swings into action , while the second peak is due to changes in τstalled , due to the action of Pol V . τmoving and τstalled both affect RecA* level in the same way , being symmetrically placed in the denominator of Equation 1 , but are influenced by different mechanisms ., Of course , various parameters that we use in our model will vary from cell to cell in a population ., Such stochasticity plays an important role in the observed behavior , probably only for those components that are present in low numbers in the cell ., Therefore , we consider that stochasticity in the number of replication forks is likely to be the most important source of cell-to-cell variability for the SOS system ., As a default we take this number , Nf , to be 2 ., However , for comparing with data obtained from cell populations , we averaged several runs where Nf was allowed to vary between 1 and 3 ( see Figures 4 and 8 ) ., Another component present in a relatively low concentration is UmuC , a variation of which is shown in Figure 7B ., Figure 7C shows that the Pol V profile is quite sensitive to ClpX ., Therefore , this might be another source of variability ., As more directly observable predictions of our model , we offer the following:, ( i ) Overexpression of ClpX should considerably reduce the Pol V concentration ., At the other extreme , the absence of ClpX would lead to Pol V being turned off at a later time than in wild-type cells ( see Figure 7C ) ., ( ii ) Overexpression of UmuC results in a flatter Pol V profile ( see Figure 7B ) , while a UmuC mutant should not be able to produce Pol V and hence should behave like the ΔUmuDC and uncleavable UmuD mutants studied in 10 ., ( iii ) We find that some overexpression of UmuD ( up to

Introduction, Results, Discussion, Materials and Methods

Escherichia coli bacteria respond to DNA damage by a highly orchestrated series of events known as the SOS response , regulated by transcription factors , protein–protein binding , and active protein degradation ., We present a dynamical model of the UV-induced SOS response , incorporating mutagenesis by the error-prone polymerase , Pol V . In our model , mutagenesis depends on a combination of two key processes: damage counting by the replication forks and a long-term memory associated with the accumulation of UmuD′ ., Together , these provide a tight regulation of mutagenesis , resulting , we show , in a “digital” turn-on and turn-off of Pol V . Our model provides a compact view of the topology and design of the SOS network , pinpointing the specific functional role of each of the regulatory processes ., In particular , we suggest that the recently observed second peak in the activity of promoters in the SOS regulon ( Friedman et al . , 2005 , PLoS Biology 3 ( 7 ) : e238 ) is the result of positive feedback from Pol V to RecA filaments .

Ultraviolet light damages the DNA of cells , which prevents duplication and thereby cell division ., Bacteria respond to such damage by producing a number of proteins that help to detect , bypass , and repair the damage ., This SOS response system displays intricate dynamical behavior—in particular the tightly regulated turn-on and turn-off of error-prone polymerases that result in mutagenesis—and the puzzling resurgence of SOS gene activity 30–40 min after irradiation ., In this paper , we construct a mathematical model that systematizes the known structure of the SOS subnetwork based on experimental facts , but which remains simple enough to illuminate the specific functional role of each regulatory process ., We can thereby identify the interactions and feedback mechanisms that generate the on–off nature of mutagenesis .

eubacteria, computational biology

journal.pgen.1006222

KdmB, a Jumonji Histone H3 Demethylase, Regulates Genome-Wide H3K4 Trimethylation and Is Required for Normal Induction of Secondary Metabolism in Aspergillus nidulans

Chromatin is the natural substrate for all eukaryotic nuclear processes such as transcription , replication , recombination or DNA repair ., Chromatin structure is necessarily dynamic and the underlying mechanisms involve remodeling of nucleosomes as well as depositing and removing posttranslational modifications on N-terminal and central residues of histones proteins ( HPTMs ) present in the nucleosome octamer 1–4 ., Some of these histone marks , such as acetyl groups on lysines , profoundly influence the chromatin landscape by neutralizing the positive charge of histones thereby weakening the interaction between nucleosomes and DNA and increasing chromatin accessibility 5 ., HPTMs also work indirectly by providing binding sites for chromatin-associated proteins that promote or inhibit specific genomic functions ., Notably , many HPTMs recruit additional chromatin-modifying enzymes that add new or remove existing marks , enabling cells to dynamically regulate chromatin structure in response to environmental or developmental cues ., Fungi have served as model systems for chromatin studies and in many basic mechanisms they are similar to higher eukaryotes but in some aspects they are quite different and this fact allows evolutionary insights into the development of chromatin regulatory systems ( reviewed in 6–8 ) ., For example , there is ground-laying work from the filamentous ascomycete Neurospora crassa , where the molecular machinery relating heterochromatin formation and DNA methylation was deciphered 9–12 ., Similar to animals also in N . crassa Heterochromatin Protein 1 ( HP1 ) , docks on di- or trimethylated lysine-9 on histone H3 ( H3K9me2/3 ) to promote heterochromatin formation 13 , 14 and in addition is important to maintain H3K27me3 , another repressive mark , at facultative heterochromatin 15 , 16 ., This mark was found to span 6 . 8% of the fungal genome 17 corresponding to over 700 transcriptionally repressed genes , some of which are upregulated upon deletion of the H3K27 methyltransferase 16 , 17 ., While H3K27 methylation and elements of Polycomb Repressive Complex 2 ( PRC2 ) responsible for depositing this mark are present in Neurospora and the Fusarium group of fungal pathogens ( see below ) this silencing mechanism has not been detected in Aspergillus species 18 ., In addition , DNA methylation has not been found in the Aspergilli although a cytosine methyltransferase is functionally expressed in A . nidulans and has a role in regulating sexual development 19 ., Mycotoxins , antibiotics , pigments and other low molecular weight natural products are summarized under the term of secondary metabolites ( SMs ) ., The Fusarium and Aspergillus genera are large groups of fungi comprising important plant and animal pathogens and they all produce ( SMs ) at certain developmental stages or under conditions of growth restriction , nutrient limitation and environmental stress ( reviewed in 20–23 ) ., It was shown initially in Aspergillus nidulans by genetic analysis that expression of the corresponding SMs biosynthetic genes , which are usually organized in gene clusters , is under chromatin control ( reviewed in 24 ) ., Under conditions of active growth SMs genes are silenced by H3 deacetylation 25 , 26 as well as by the H3K9 methylation machinery of ClrD ( KMT1/ DIM-5 homolog ) and the hpo homolog HepA 27 ., Interestingly , H3K4 methylation and a subunit of the COMPASS complex which are usually known to be associated with gene activation , also contribute to silencing although this has only been observed for a small subset of SM genes 28 ., Several recent studies in a number of other fungi have implicated heterochromatin as a regulator of secondary metabolism and the production of virulence factors ., In the plant pathogens F . graminearum ( wheat and maize pathogen ) and F . fujikuroi ( rice pathogen ) as well as in the fungal endophyte Epichloë festucae , H3K9me3 and H3K27me3 regulate expression of specific gene clusters responsible for the production of secondary metabolites 20 , 23 , 29–32 ., H3K9me3 and HP1 were also shown to negatively regulate other virulence factors such as genes encoding small secreted proteins ( SSPs ) in Leptosphaeria maculans 29 ., How HPTM patterns change as SM clusters switch from a repressed state to an active state is not completely understood ., The requirement of histone H3 and H4 acetylation for SM gene expression is well documented in Aspergillus species through HDAC inhibitor studies and SAGA- complex mutants 33–35 ., Interestingly , co-cultivation of A . nidulans cells with Streptomyces rapamycinicus led to an anomalous activation of several SM genes in the fungus 36 and this process is correlated with increased H3 acetylation of the corresponding genes and strictly dependent on GcnE , the catalytic subunit of the A . nidulans SAGA acetylation complex 37 ., Also in F . fujikuroi , activation of the GA , bikaverin and fumonisin clusters was correlated with increased acetylation of H3K9 38 ., In contrast to acetylation , the role of histone methylation in fungal SM gene expression is much less clear ., In F . graminearum , silent SM clusters are highly enriched for repressive H3K27me3 , whereas trimethylated H3 lysine 4 ( H3K4me3 ) , an activating mark , is apparently excluded ., Upon deletion of the H3K27 methyltransferase kmt6 , the silent fusarin C and carotenoid clusters are activated , but H3K4me3 does not accumulate in these clusters 30 ., A similar situation was shown in F . fujikuroi where increases in H3K4me2 were only observed in two genes of the gibberellin ( GA ) cluster ., Similar to the case for H3K4me , expression of SM cluster genes in F . graminearum was not associated with increased H3K36me3 30 ., In contrast , H3K36me3 was gained for the sterigmatocystin ( ST ) and several other SM gene clusters in A . nidulans during activation 18 , 39 , 40 ., H3K4me3 is an HPTM with important roles in transcription and this mark is generated by the COMPASS ( Complex associated with Set1 ) protein complex containing the Set1 methyltransferase catalytic subunit in addition to several regulatory and scaffold proteins 41 ., COMPASS is not essential in A . nidulans although synthetic lethality of Set1 and Swd1 subunits was found with mutations in mitotic regulators 42 ., Generally , H3K4me3 has been shown to be recognized by three different domains associated with proteins of various functions ., One recognition module is the PHD domain , present for example in the “Inhibitor of Growth” ( ING ) protein , which recruits histone acetyltransferase ( HAT ) and deacetylase ( HDAC ) complexes 43 , 44 ., H3K4me3 is also recognized by the double TUDOR domain of JMJD2A , a JmjC family demethylase that removes methyl groups from di- or trimethylated H3K9 45 and by the tandem chromodomain of CHD1 , an ATP- dependent nucleosomal remodeler 46 recently shown to be necessary for inhibition of intragenic initiation or initiation from cryptic promoters and thus maintaining normal transcript elongation 47 ., Accordingly , H3K4me3 plays a central role in the chromatin regulatory network ., Usually , H3K4me3 peaks at the transcription start sites ( TSSs ) and its occurrence is correlated with gene expression 48 ., However , the Set1 protein also displays some moonlighting activities as it recruits deacetylase activity independently from the H3K4me3 mark and subsequently promotes heterochromatin formation and transcriptional repression at distinct loci in the fission yeast genome 49 ., This evidently negative role of the COMPASS was also documented for regulation of SMs production in three different Aspergillus species carrying genetically engineered COMPASS mutations 28 , 50 , 51 ., Silencing specific SM gene clusters might be related to previously documented subtelomeric silencing functions of the COMPASS complex 41 and mechanistically similar to the recently identified heterochromatin-promoting role in fission yeast 49 ., Dynamic demethylation of lysine residues adds additional complexity to the modulation of transcription by lysine methylation 3 , 52 ., Recently we showed that KdmA , a JMJD2/JHDM3 family H3K9/36me3 demethylase 53 , 54 can , in equal measure , positively and negatively influence gene expression in A . nidulans 18 ., Here , we characterize another member of the JmjC demethylase family , KdmB , which acts on H3K4me3 in vivo , thus is assigned to the Jarid group of enzymes ., Jarid ( JMJ–AT-rich interacting domain-containing protein ) subfamily demethylases have been shown to target di- and trimethylated H3K4 and are therefore generally considered to be repressors of gene transcription , though they can also act as activators 55 ., For example the function of mammalian RBP2 ( retinoblastoma binding protein 2 , alias JARID 1A or KDM5A according to the new nomenclature 56 ) in transcription regulation is context dependent ., RBP2 represses transcription via H3K4me3 demethylation and association with an HDAC complex , however when associated with retinoblastoma protein ( pRb ) , it activates certain genes in the mammalian genome 57 ., Similarly , the D . melanogaster ortholog LID can repress transcription via H3K4me3 demethylation , however when associated with the MYC transcription factor , its demethylase activity is inhibited and consequently the LID-MYC complex mediates gene activation 58 , 59 ., These examples demonstrate that Jarid demethylases can act directly on their target genes in a context dependent positive or negative manner ., In this work we studied the Jarid-type demethylase in A . nidulans by reverse genetics and performed genome-wide HTPM profiling by mass spectrometry of histones , by ChIP analysis of H3K4me3 , H3K9me3 , H3K36me3 and H3 acetylation on K9 and K14 ( H3Ac ) modifications in wild type and compared the results with the KdmB mutant ., We recorded these HPTM changes in parallel with the transcriptome under optimal physiological conditions promoting active growth ( primary metabolism ) as well as under stationary-phase conditions that lead to SM production ( secondary metabolism ) ., Comparison of ChIP-seq profiles with RNA-seq of the same cultures allowed us to correlate transcriptional changes with changes in chromatin landscapes across different conditions and genetic backgrounds ., Histone proteomic analysis in wild type and the KdmB histone H3K4 demethylase mutant provided direct evidence for H3K4me3 as the dominant substrate for KdmB and confirmed that A . nidulans does not feature H3K27me3 , the canonical facultative heterochromatic mark in other eukaryotes and responsible for SM gene silencing in a number of other fungi ., Based on the domain composition of the full length KdmB ( AN8211 ) and detailed analysis of the amino acid sequences of the catalytic JmjC domains of histone demethylases from yeast to humans , KdmB was classified as a Jarid1-type histone H3 lysine 4 demethylase ( Fig 1 ) ., Residues responsible for substrate recognition of Jarid demethylases are not known due to the lack of available crystallographic data , although the conserved amino acids required for substrate recognition in the JMJD2 subfamily of lysine K9 and K36 histone H3 demethylases ( marked in green in Fig 1A ) are not present in the Jarid group 60 ., Domain analysis revealed that KdmB is more similar to the proteins from higher eukaryotes than from budding yeast ., Specifically , we found that KdmB contains a putative ARID/Bright domain and a C5-HC2 zinc finger motif and an additional PHD domain at the C-terminus , which are both absent from the budding yeast homolog ( Fig 1B ) ., To investigate the in vitro specificity of KdmB we heterologously expressed KdmB as a GST fusion protein in E . coli ., KdmB has predicted molecular weight of 216 kDa but the resulting full size recombinant protein was not sufficiently soluble ., Another construct producing a truncated KdmB protein without the second PHD domain , however , was readily soluble under native buffer conditions ., This KdmB fusion containing residues 1 to 922 displayed an apparent mass of roughly 130 kDa ( S1A Fig ) ., In vitro demethylase assays ( DeMt ) were subsequently performed with purified GST-KdmB ( 1–922 ) and calf thymus histones as a substrate ., Products of the DeMt reactions were detected with modification-specific antibodies by Western blot ( S1B Fig ) ., Under our assay conditions , we found a decrease in trimethylation signals for all three tested lysine residues ( H3K4me3 , H3K9me3 and H3K36me3 ) and the strongest reduction in abundance was seen in H3K9me3 ., Acetylation was not reduced by the enzyme , as expected ., Consistent with KdmB being a JmjC-type demethylase , the activity of the GST-KdmB ( 1–922 ) , fusion protein was dependent on the presence of the cofactors α- ketoglutarate and Fe2+ ( S1B and S1C Fig ) ., In our assay conditions we observed high standard deviations between independent replicates of H3K4me3 and H3K36me3-specific Westerns ., This could be due to experimental variation in enzymatic activity of different batches of the purified recombinant enzyme ., The very broad substrate range of KdmB in vitro is unexpected for this Kdm5-family member because so far the identified and tested enzymes target either H3K4me2/3 ( Jarid1 group enzymes ) or H3K9/36me2/3 ( Jmjd2 group ) ., However , it is possible that the absence of PHD-finger 2 , interacting proteins or the presence of the GST domain compromises substrate specificity in our assay ., Although none of the KdmB orthologs identified so far demonstrated such broad substrate specificity in vitro 57 , 63–66 , the in vitro demethylase activity found in our assays suggests that this protein possesses histone demethylase activity ., To determine whether KdmB can act as a histone demethylase in vivo , we performed LC- MS/MS on acidic extracted histones from actively growing A . nidulans wildtype and kdmBΔ cells ( see Materials and Methods for description of gene deletion procedure ) ., In wildtype , mass spectrometry revealed that 71 . 3% of H3K4 peptides contain at least one methyl-group at the K4 position ., We detected all three forms of methyl-H3K4 peptides and found that H3K4me3 is the most abundant ( 47 . 5% of total H3K4 peptides ) , followed by di-methylated ( 13 . 5% ) and mono-methylated H3K4 ( 10 . 3% ) ( Fig 2 ) ., Notably , our measurements revealed an almost 20% increase in global H3K4 trimethylation in the kdmBΔ strain ( 57% H3K4me3 ) ., Because the levels of H3K4me2 , H3K4me1 and unmodified H3K4 were concomitantly decreased in the mutant in roughly the same range as H3K4me3 increased we concluded that in vivo KdmB primarily acts to demethylate H3K4me3 ., The MS results also revealed that in vivo KdmB does not target H3K36me3 as these levels remained constant in histones of kdmBΔ cells ( S2 Fig ) ., Interestingly , the overall low marking of H3K9 by trimethylation ( 1 . 53% of the mapped peptides ) was further reduced ( to 0 . 2% of the mapped peptides ) in the mutant ., This contrasts the in vitro assay results which showed a strong H3K9me3 demethylating activity of recombinant KdmB ( S1 Fig ) ., The further reduction of H3K9me3 in kdmBΔ cells might be attributable , however , to an increase in the opposing , positively acting H3K4me3 mark limiting the possibility to deposit or maintain H3K9 trimethyl marks in the target regions ., Strikingly , in vivo , global H3 N-terminal lysine acetylation ( H3K9ac/K14ac ) was increased almost by 20% in the kdmBΔ strain at the expense of unmodified peptides of H3 which are reduced from 22% in the wild type to 6% in the mutant ( S2A Fig ) ., This more abundant histone acetylation could be the consequence of both stronger marking by acetylases and/or reduced deacetylation ., The latter mechanism has already been reported in connection with KdmB homologs in mammals where RBP2 ( Jarid1a ) and PLU1 ( Jarid1b ) recruit the Rpd3S histone deacetylase complex 57 , 67 ., Altogether , our data demonstrate H3K4me3 demethylation activity of KdmB in A . nidulans cells and lack of this activity in kdmB deletion cells leads to a shift in modification equilibrium with more abundant positive ( H3K4me3 , H3Ac ) and less negative ( H3K9me3 ) marks ., To determine the genomic regions in which KdmB influences H3K4me3 levels we performed genome-wide ChIP analysis ( ChIP-seq ) in wild type and kdmBΔ strains with antibodies specific to H3K4me3 28 ., As our global histone analysis revealed a crosstalk of this modification to H3K9 trimethylation as well as to H3K9/K14 acetylation , we also included these marks in ChIP-seq ., Although no changes occurred for H3K36 trimethylation at the level of bulk histones between WT and the kdmB mutant , we were interested if locus-specific differences occur and thus analyzed also this mark by ChIP-seq ., As previous studies from our lab and by others revealed a crucial function of chromatin structure and histone modifications on the regulation of secondary metabolite biosynthesis ( SMB ) , we performed all subsequent RNA-seq and ChIP-seq experiments not only under the already described standard active growth conditions representing primary metabolism ( PM; 17h liquid shake cultures , no nutrient limitation ) but also under conditions promoting secondary metabolism ( 48h liquid shake cultures , nutrient depletion , see S3 Fig ) ., To monitor the distribution of the tested chromatin modifications along A . nidulans genes , we used chromosome IV as an example and plotted the wild type distribution of H3K4me3 , H3K36me3 , and H3K9/14ac across the promoters and open reading frames ( ORFs ) of all genes on this chromosome ( Fig 3A ) ., In this analysis , all genes are aligned to the predicted ATG ( position 0 ) and read counts per million of mapped reads ( CPM ) are analysed in a 2 kb window starting with 500 bp of their 5´UTR and promoter sequences ( -500 ) followed by 1500 bp of their coding region ., This revealed that the pattern of modifications reflects the distribution observed in other model organisms including fungi 30 , 68–70 ., H3K4me3 was enriched in characteristic peaks spanning the first three nucleosomes ( around 500 bp ) of the coding region , whereas H3K36me3 was enriched near the 3’ regions of genes ., Finally , H3 acetylation was enriched in the promoter , with highest levels apparent in the first nucleosome just downstream of the predicted translation start sites ., To explore the general relationship between H3K4me3 and transcription we quantified the average level of H3K4me3 in a 2 kb window around the predicted start codon of each gene ( average CPM from -500 to +1500 ) and related this value to the average expression level ( expressed as RPKM , reads per kilobase per million reads ) of the corresponding gene in both culture conditions ( PM and SM ) ., In the resulting scatterplot ( Fig 3B ) two groups of genes became apparent , i . e . those that displayed high levels of H3K4me3 ( log2 RPKM>5 ) and a second group that showed low to no H3K4 trimethylation ( log2 RPKM≤5 ) ., Correlation of H3K4me3 levels with transcription of the corresponding gene revealed an overall positive correlation between H3K4me3 levels and transcript abundance ( Fig 3B ) ., This suggests that , similar to other well-studied models , H3K4 trimethylation is a marker for actively transcribed genes ., To better characterize the function of KdmB in the context of transcriptional regulation we next compared by ChIP-seq the distributions of four histone modifications in wild type and kdmBΔ ( Fig 4 ) under active growth conditions ( PM ) and during SM ., The kdmB deletion did not cause any gross phenotypic changes in the mutant strain which was rather similar to the wild type in growth rates and nutrient consumption ( S3 Fig ) ., ChIP-seq combined with RNAseq analysis revealed the H3K4me3 enriched domains which coincide with transcriptional activity ., In the example shown in Fig 4 we noticed , on the gross genomic scale , an overlap between the positively acting marks H3K4me3 , H3K36me3 and H3Ac ., In contrast , repressing H3K9me3 marks are enriched mainly in pericentromeric and subtelomeric regions and a few isolated H3K9me3 blocks exist ( on the left arm of chromosome IV , for example ) ., At the gross genomic scale the comparison of the chromatin landscape for H3K4me3 marks in chromosome IV between actively growing ( 17 h cultures ) wild type and kdmBΔ cells did not reveal any obvious changes ., Moreover , at this scale , no large domains were visibly changed for the other tested modifications ( H3Ac , H3K36me3 , H3K9me3 ) ., Because our mass spectrometry analyses uncovered increased H3K4me3 and H3Ac in the mutant , we reasoned that changes in the levels of these histone marks must occur at a subset of individual genes ., To test this , we analyzed H3K4me3 levels in genes that were differentially expressed between wildtype and kdmBΔ ., We first examined genes with low H3K4me3 levels ( log2 ( RPKM ) ≤ 5 and found that 301 genes displayed higher expression levels in the wildtype ( WT-up/Group 1 , Fig 5A ) suggesting that for this group KdmB is required for normal expression levels ., In contrast , 501 genes had higher expression in kdmBΔ ( kdmBΔ-up/ Group, 2 ) which points to a repressing function of the protein in these loci ., In the gene set featuring high H3K4me3 levels ( log2 ( RPKM ) > 5 we again identified both up- and down-regulated genes; 455 genes were expressed at higher levels in wild type ( WT-up/Group, 3 ) and 133 genes were expressed at higher levels in kdmBΔ ( kdmBΔ-up/ Group 4 ) ., The analysis showed that KdmB influences transcriptomes in both directions ., For around 750 genes KdmB function is necessary for normal transcription , whereas for around 630 genes KdmB has a negative function ., The repressive role of KdmB was found in both categories , i . e . on genes carrying low ( kdmBΔ-up/G2 ) or high ( kdmBΔ-up/G 4 ) H3K4me3 levels ., Significantly , the group with normally low H3K4me3 ( G2 ) displayed a marked increase in this histone mark in the kdmBΔ mutant concomitantly with increased transcript levels ., One representative of this group is shown in Fig 5C for a gene ( locus AN6321 ) which is basically not transcribed in the wild type but which gains both positive marks and transcripts in the kdmBΔ strain ., Although we have not tested this directly , the strict correlation between increased H3K4me3 levels and transcription , along with the in vitro K4me3-demethylase activity of KdmB , suggests that at least some of these loci are direct targets of KdmB ., A slightly different situation was found for the second gene set highly decorated with H3K4me3 ., Although a subset of these genes showed increased expression in the kdmBΔ mutant ( kdmBΔ-up/G 4 ) , this was not accompanied by an increase in H3K4me3 probably due to the already very high K4 methylation levels in the wild type ., Consequently , a further increase would hardly be possible and thus the effect of kdmB deletion on H3K4 trimethylation is more subtle compared to genes generally not heavily marked by H3K4me3 ., In contrast to the repressive function , KdmB also seems to have a positive role in transcription ., kdmB deletion led to reduced expression of 750 genes belonging to both low ( WT-up/G1 ) or high ( WT-up/G3 ) H3K4me3 groups , accompanied by lower H3K4me3 , on average , in the mutant ., Based on these correlations we can conclude that KdmB function is required for normal expression of these roughly 750 genes , but whether KdmB directly targets these loci or indirectly affects transcription via the transcriptome network remains to be determined ., We also constructed metaplots of H3K4me3 distributions under SM conditions ( S4 Fig ) ., Under these growth conditions a similar correlation was observed , i . e . H3K4me3 levels were reduced in genes that were downregulated in kdmBΔ , whereas the genes upregulated in the mutant showed no drastic change ( in the high H3K4me3 group ) or somewhat higher H3K4 trimethylation ., However , in locus-specific analysis by RNA-seq and ChIP-seq ( see below ) , we also found some transcriptionally silent regions with high H3K4me3 as well as some highly transcribed genes with very low levels of this mark ( see analysis below ) indicating that specific genomic regions exist in which this general positive correlation between H3K4me3 and transcriptional activity does not apply ., Our initial correlation analysis of H3K4me3 and transcription revealed that among genes requiring KdmB for full transcription , the category of SMB genes was significantly enriched ( p < 0 . 05 ) ., In further analysis , PM and SMB genes were separated based on functional categories and this bioinformatic approach created a large group of genes ( 5676 genes ) predicted to be involved in general cellular functions and metabolism ( category “cell structure and function” abbreviated CSF ) and a smaller group of 149 genes predicted to be involved in SMB ( category “SM clusters” ) ., 71 , 72 ., Fig 6 shows that under PM conditions , approximately 5% of genes involved in CSF and 15% of genes assigned to SMB were affected by the kdmB deletion ., The majority of A . nidulans SM cluster genes are not under/ during PM conditions , thus it is not surprising that differential expression of SM genes is largely restricted to the 48h cultures ., Interestingly , several genes belonging to a gene cluster with a so far unidentified product were highly upregulated in the mutant at this 17h time point and this transcriptional pattern will certainly facilitate the future identification of the product derived from this predicted SM cluster ., In contrast to the mild effect on SM gene expression during PM conditions , KdmB-deficient cells showed significantly altered patterns of gene expression when cells were collected from cultures under SM conditions ., Over 50% of all predicted SM genes were misregulated in the mutant ., The majority of these displayed lower expression , while approximately 10% of SM genes showed higher expression in the kdmBΔ strain ( Fig 6A , upper panel ) ., In contrast , during the same culture condition only ~10% of genes not involved in SM were differentially transcribed in kdmBΔ ., These data demonstrate that KdmB is required for normal induction of the majority of SM clusters in A . nidulans ., It is probably relevant to note that the defect in SM cluster activation in the kdmB mutant is not due to a lack of wide-domain activator expression as laeA , veA , velB and velC are normally transcribed in the mutant ( changes between WT and kdmBΔ log2 ≤ ± 1 , 7 ) ., The lower panel of Fig 6A presents the number of deregulated genes within each category and time point ., During primary metabolism ( 17h ) KdmB function is required for a relatively small number of genes ( 143 genes in CSF and 10 genes in SM ) ., In contrast , in the nutrient limited 48h cultures gene expression profiles are changed considerably in the mutant: 598 genes ( 401 CSF and 97 SMB genes ) require KdmB function for normal expression and 569 genes ( 547 CSF and 22 SMB genes ) are negatively influenced by the regulator ., These data suggest that KdmB is primarily required during the stationary phase and obviously plays an important role for the expression of the majority ( 97 out of 149 of genes involved in SMB We also tested whether transcriptional changes in kdmBΔ were correlated with changes in SMB biosynthesis ., For this we performed HPLC-MS/MS analyses of cultures grown in two different media , i . e . in conventional minimal medium used throughout the studies ( AMM ) and in a specialized SM-promoting ZM medium ( see Materials and Methods section ) ., The comparison of WT and mutant culture extracts , grown in AMM medium , revealed a strongly decreased production of sterigmatocystin and emericellamides C and D ( Fig 6B , left chromatograms ) but other metabolites such as emodin and its derivatives were increased in kdmBΔ ( Fig 6B , chromatograms a and c ) ., However , our RNA-seq data showed that genes encoding for enzymes involved in emodin biosynthesis embedded in the mdpL-A monodictyphenon pathway are not differentially expressed between WT and the kdmB mutant ( S13 Fig ) ., To accommodate these differences , we speculate that the decreased transcription of other secondary metabolite clusters , such as the sterigmatocystin cluster , may lead to higher levels of available emodin precursors , such as acetyl-CoA and malonyl-CoA , and thereby to an increased synthesis of emodin derivatives ., ZM culture extracts revealed reduced levels of orsellinic acid in kdmBΔ ( Fig 6B , right chromatograms ) , consistent with our RNA-seq data showing a decreased expression from the orsellinic acid gene cluster in the kdmB deletion ( S10 Fig ) ., The complete list of identified metabolites together with LC-MS and LC-MS2 data are shown in the S3 Table ., We also carried out correlation analyses between H3 acetylation and H3K4 methylation in genes which are differentially regulated in the kdmB mutant ( S5 Fig ) ., For those genes where KdmB is required for full expression and which are consequently higher transcribed in the wild type ( categories WT-up/G1 and G3 ) H3 acetylation levels are also higher , independently of H3K4 trimethylation ., The same is true for genes which are negatively influenced by KdmB ( kdmBΔ-up/G2 ) but only if H3K4me3 levels are low ., On the contrary , genes with high H3K4me3 levels under negative KdmB influence ( kdmBΔ-up/G4 ) , acetylation levels are lower than in the wild type despite higher expression of the corresponding genes in this group ., The molecular basis of this effect has not been investigated further in this study but it would certainly be interesting to determine if KdmB impacts acetylation indirectly or directly through protein interactions with HDACs or HATs ., We also examined a possible influence of KdmB on the distribution of H3K36me3 in genes expressed under primary metabolic conditions ( S6 Fig ) ., We have previously shown that this mark is associated with active transcription and that , at some tested loci , the trimethylated H3K36 state is removed by KdmA , another A . nidulans JmjC-containing protein belonging to the KDM4 family 18 ., The vast majority of A . nidulans genes are highly decorated by this mark under PM conditions ( around 9 , 100 genes ) ., We did not observe significant differences in the levels or in the distribution of this mark in the kdmBΔ strain neither in this group nor in the group carrying low H3K36me3 levels ( 1249 genes ., This indicates that KdmB is not a demethylase of trimethyl-H3K36 in vivo ., Around 13% of the 9 , 100 genes are de-regulated in the kdmB mutant strain but despite this differential expression there are no significant differences in the associated H3K36 trimethylation levels ., This means that , at least for the gene set in which KdmB influences transcription , it does not do this via manipulating H3K36me3 levels ., The genome-wide distribution pattern of H3K9me3 supports the previously reported low levels of H3K9 trimethylation in A . nidulans wild type cells where we found approximately 1 . 5% of peptides carrying this mark ., 18 ., In ChIP-seq , the H3K9me3 pattern correlates with AT-rich domains flanking the subtelomeric regions but also includes sites along the chromosome arms , as shown on the left arm of chromosome IV ( Fig 4 ) ., Inspection of H3K9me3-associated regions revealed that many SMB gene clusters such as the penicillin ( S7 Fig ) , sterigmatocystin ( S8 Fig ) , austinol ( S9 Fig ) , orsellinic acid ( S10 Fig ) and terrequinone A ( S11 Fig ) are flanked by H3K9me3 domains at either one ( e . g . the ST and TDI clusters ) or at both sides ( e . g . the PEN cluster ) of the cluster ., Whether these structures are functionally relevant for the regulation of SM gene clusters remains obscure but possible since deletion of the H3K9 methyltransferase gene clrD or of hepA , the gene coding for the protein recognizing H3K9me3 , lead to up-regulation of genes within these clusters 27 ., The observation that many H3K9me3 blocks are found in close proximity to SMB gene clusters raises the possibility that higher

Introduction, Results, Discussion, Materials and Methods

Histone posttranslational modifications ( HPTMs ) are involved in chromatin-based regulation of fungal secondary metabolite biosynthesis ( SMB ) in which the corresponding genes—usually physically linked in co-regulated clusters—are silenced under optimal physiological conditions ( nutrient-rich ) but are activated when nutrients are limiting ., The exact molecular mechanisms by which HPTMs influence silencing and activation , however , are still to be better understood ., Here we show by a combined approach of quantitative mass spectrometry ( LC-MS/MS ) , genome-wide chromatin immunoprecipitation ( ChIP-seq ) and transcriptional network analysis ( RNA-seq ) that the core regions of silent A . nidulans SM clusters generally carry low levels of all tested chromatin modifications and that heterochromatic marks flank most of these SM clusters ., During secondary metabolism , histone marks typically associated with transcriptional activity such as H3 trimethylated at lysine-4 ( H3K4me3 ) are established in some , but not all gene clusters even upon full activation ., KdmB , a Jarid1-family histone H3 lysine demethylase predicted to comprise a BRIGHT domain , a zinc-finger and two PHD domains in addition to the catalytic Jumonji domain , targets and demethylates H3K4me3 in vivo and mediates transcriptional downregulation ., Deletion of kdmB leads to increased transcription of about ~1750 genes across nutrient-rich ( primary metabolism ) and nutrient-limiting ( secondary metabolism ) conditions ., Unexpectedly , an equally high number of genes exhibited reduced expression in the kdmB deletion strain and notably , this group was significantly enriched for genes with known or predicted functions in secondary metabolite biosynthesis ., Taken together , this study extends our general knowledge about multi-domain KDM5 histone demethylases and provides new details on the chromatin-level regulation of fungal secondary metabolite production .

In this work we monitored by proteomic analysis and ChIP-seq the genome-wide distribution of several key modifications on histone H3 in the model fungus Aspergillus nidulans cultivated either under optimal physiological conditions ( active growth ) or less favourable conditions which are known to promote the production of secondary metabolites ( SM ) ., When we correlated the chromatin status to transcriptional activities in actively growing cells we found that the silenced SM gene clusters are flanked by heterochromatic domains presumably contributing to silencing but that the bodies of the clusters only carry background levels of any of the investigated marks ., In nutrient-depleted conditions , activating marks were invading some , but by far not all transcribed clusters , leaving open the question how activation of these regions occurs at the chromatin level ., Surprisingly , a large number of these gene clusters actually depend on KdmB for normal activation and it will be interesting to see in future how this protein thought to mainly act as repressor by removing positive H3K4m3 marks switches gears to activate transcription directly or indirectly .

aspergillus, fungal genetics, gene regulation, dna-binding proteins, dna transcription, aspergillus nidulans, fungi, model organisms, epigenetics, chromatin, research and analysis methods, mycology, genomics, chromosome biology, proteins, gene expression, chemistry, molds (fungi), histones, biochemistry, fungal genomics, cell biology, post-translational modification, acetylation, genetics, biology and life sciences, yeast and fungal models, physical sciences, chemical reactions, organisms

journal.ppat.1006940

Pleiotropic roles of Clostridium difficile sin locus

Clostridium difficile , a major nosocomial pathogen , is the causative agent of antibiotic-associated diarrhea and pseudomembranous colitis 1 , 2 ., Every year , nearly half a million cases of C . difficile infections ( CDI ) occur in the United States and result in approximately 14 , 000 deaths 3 ., C . difficile toxins damage the colonic epithelium , which results in moderate to severe diarrhea 4 ., Recent studies have shown that these toxins are essential for C . difficile pathogenesis 4–7 ., Due to the strictly anaerobic nature of the vegetative cell , C . difficile survives outside the host in the form of dormant spores , which are highly resilient and resistant to most disinfectants ., Thus , C . difficile spores are critical for its host to host transmission and persistence in the hospital environment 8 ., C . difficile Toxins A and B are encoded by the tcdA and tcdB genes respectively , and their expression is dependent on TcdR , an alternative RNA polymerase sigma factor 9–11 ., Environmental stresses , such as alteration of the redox potential , high temperature , or limitation of nutrients like glucose , and biotin , modulate toxin production by influencing the expression of tcdR 9–12 ., Similar to toxin production , the sporulation pathway in C . difficile is also known to be influenced by nutrient availability and uptake 13 , 14 ., The regulators involved in controlling toxin synthesis in response to nutrients are the global regulatory proteins CcpA and CodY 14–18 ., Among them , CcpA mediates glucose-dependent toxin gene repression 15 , 16 , and CodY blocks the transcription of toxin genes during the exponential growth phase of the bacterial culture 17 , 18 ., Other than affecting toxin production , mutations in codY and ccpA were also found to affect sporulation 13 , 16 ., Other genes that are known to influence both toxin production and sporulation include spo0A , sigH , and rstA 19–22 ., New evidence suggests that the toxin , motility , and sporulation regulatory networks are linked together in C . difficile 19 , 23 , 24 ., The sigma factor SigD needed for transcription of the flagellar operon was identified to regulate tcdR transcription to influence toxin production 25 , 26 positively ., Mutations in spo0A , rstA , and sigH also influenced motility along with toxin production and sporulation 19–22 ., This study identified that mutation of the sin locus in C . difficile could affect toxin production and sporulation along with motility and thus reports a new regulatory element of this network ., In Bacillus subtilis , the sin ( sporulation inhibitor ) locus codes for two proteins SinR and SinI and regulates several genes involved in sporulation , motility , competency , proteolysis , and biofilm formation 27–31 ., In this study , we have created C . difficile sin locus mutants in two different strains ., Using RNA-Seq analysis , we compared the transcriptome of the mutants with respective parent strains to identify and assess the transcriptional regulation of sin locus coded regulators ., Follow up phenotypic analyses and complementation experiments showed that the Sin regulators in C . difficile are also pleiotropic as in B . subtilis ., Here , their regulatory roles in toxin production , sporulation , and motility were further investigated and discussed ., In B . subtilis , the sin locus carries two small ORFs , sinI and sinR 32 , 33 ( Fig 1A ) ., B . subtilis SinR ( BsSinR ) is a DNA-binding protein that binds to a conserved DNA sequence upstream of the translational start site of target genes to negatively control their transcription ., SinI , encoded by a gene adjacent to sinR , has an antagonistic relationship with SinR and binds directly to the SinR protein to inhibit its activity ., This causes the pathways that were repressed by SinR to switch on ., In B . subtilis , SinR contains 113 aa , and the DNA binding domain is located at the N-terminus part , which spans from residues 5–61 32 , 33 ( Fig 1A ) ., The C-terminal part of SinR forms alpha-helices and is responsible for multimerization and SinI interaction ., The SinI protein , on the other hand , resembles a truncated SinR without the DNA binding region and carries only the alpha-helical structure to drive the hetero-dimerization of SinR-SinI complex 32–34 ., In C . difficile the sin locus contains two ORFs CDR20291_2121 and CDR20291 _2122 ( in C . difficile R20291 reference genome ) , which codes for proteins that are 43% and 35% identical to B . subtilis SinR , respectively ( Fig 1B ) ., Both these proteins are predicted to be DNA-binding since they carry HTH ( Helix-Turn-Helix ) domains in their N-terminal regions ., Hence we named CDR20291_2121 as sinR and CDR20291_ CD2122 as sinR’ ., The C . difficile SinR ( CdSinR ) contains 112 amino acids , and its predicted HTH domain spans residues 11 to 66 ., The SinR’ ( CdSinR’ ) protein carries 105 aa , and its predicted HTH domain spans from residues 7 to 62 ( Fig 1B ) ., Both CdSinR and CdSinR’ shows the highest homology to BsSinR in this DNA-binding domain , where within the 50 residues of HTH domain , 13 of them are identical and 19 of them represent conservative substitutions ( Fig 1C ) ., CdSinR and CdSinR’ shows similarity with each other ( 33% identity ) only in their N terminal DNA binding domain ., The C terminus multimerization domains of these proteins show variations , and there is less similarity of CdSinR and CdSinR’ to BsSinR and each other in this region ., In various Bacillus sp ., SinR homologs are known to control the expression of the genes adjacent to the sin loci ., Thus , identifying genes adjacent to the sin loci were helpful in predicting at least a few functions of the Sin regulators in these bacterial species ., For example , in B . subtilis , the sin locus is adjacent to the tapA-sipW-tasA operon , and SinR represses the expression of this operon whose products are involved in the production of the biofilm matrix 31 ., In Bacillus anthracis , the sin locus is next to calY that codes for camelysin , a cell surface associated protease , and SinR in this species is known to repress the calY expression 35 ., In C . difficile , the sin locus is located in between cynT ( codes for carbonic anhydrase ) and CDR20291_2123 ( unknown function ) ( Fig 1B ) and is not close to any other genes that are known to be essential for virulence in this pathogen ., Thus , the location of the sin locus in C . difficile chromosome did not provide us any clues about its possible functions ., To get more information about the locus and its role in C . difficile physiology we decided to construct and characterize mutants in sin locus ., An erythromycin resistant marker was introduced in the sinR at nucleotide 141 using Clostron , a TargeTron-based group II intron in C . difficile JIR8094 36 and R20291 strains 37 ., The presence of the retargeted intron in the correct gene in both mutant strains was confirmed by PCR ( S1 Fig ) ., In B . subtilis , three different promoters drive the transcription of the sin genes 33 ., In B . subtilis , the polycistronic sinIR transcript is produced from two different promoters , and the sinR transcript is driven from an independent promoter immediately downstream of sinI ( Fig 1A ) 33 ., In C . difficile , the operon upstream of sin locus transcribes in the opposite direction , and no read-through transcription of sin locus is possible from its promoter ( Fig 1B ) ., Using cDNA prepared from the JIR8094 and the mutant strain , we performed RT-PCR analysis and checked for the presence of sinR , sinR’ and sinRR’ transcripts ( S2 Fig ) ., We could detect sinR , sinR’ and also the read through sinRR’ transcripts , which confirmed that the sinR and sinR’ are transcribed as a single transcript ( S2 Fig ) ., When the same analysis was performed using the mutant strain cDNA both the sinR , sinR’ and sinRR’ transcripts were absent ( S2 Fig ) ., The QRT-PCR analysis of the sinR mutant showed significant reduction of both sinR and sinR’ transcript levels ( S2 Fig ) ., It also revealed that similar to B . subtilis , the C . difficile sin locus is expressed between late-exponential and early-stationary growth phase ( 10 to 12 h ) ( S2 Fig ) ., Similar results were obtained in RT-PCR analyses of cDNA from the R20291 strain ( S2 Fig ) ., When we performed the western blot analysis using the SinR and SinR’ specific antibodies ( see M&M ) , both SinR and SinR’ were found to be absent in the mutant ( S3 Fig ) ., Our western blot and the RT-PCR results together suggest that sinR and sinR’ are part of an operon ., However , there is a possibility that sinR’ could have an independent promoter coded within the sinR coding region , which was not expressed in the growth conditions tested ., Since the insertion of the intron in sinR ( first gene in the operon ) disrupted both sinR , sinR’ transcripts , and SinR , SinR’ production in the growth conditions tested , we named the mutant strains with the disrupted sinR gene as JIR8094::sinRR’ and R20291::sinRR’ ., We first analyzed the impact of sin locus inactivation on the growth of C . difficile in TY medium ., During the exponential phase of the growth , both parents and mutants grew at a similar rate ., However , when they entered the stationary phase , we observed a decrease in the turbidity of the mutant cultures as measured as OD@600 nm ( S3 Fig ) ., We performed the Triton X-100 autolysis assay to check the influence of SinRR’ on global autolysis of C . difficile 25 ., We used the 16h old stationary phase culture to perform this assay , where the R20291::sinRR’ lysed at a faster rate compared to the parent ( S4 Fig ) ., These results suggested that inactivation of sinRR’ induced autolysis in C . difficile ., In B . subtilis , SinR along with another regulatory protein SlrR represses the expression of lytA-lytB-lytC and lytF autolysins 38 ., Our initial observation of lysis phenotype in the sinRR’ mutants suggested that like B . subtilis SinR , C . difficile SinR might also be controlling the autolysin genes ., In B . subtilis the SinR is a pleiotropic regulator and controls various pathways including autolysis 29–31 , 33 , 38 , 39 ., We suspected that SinR and SinR’ in C . difficile might also regulate several targets to control multiple functions ., Hence , to identify the sinRR’ regulated pathways in C . difficile , we performed the transcriptome analysis of the sinRR’ mutants in comparison with their respective parents ., Based on the growth pattern of the sinRR’ mutants ( S3 Fig ) and the expression kinetics of sinRR’ in the parent strains ( S2 Fig ) , we decided to compare the transcriptomes of mutant strains with their respective parent strains during the early stationary phase ( i . e . , 12 h of growth ) in TY medium ., We used three biological replicates and genes were considered differentially expressed if the fold change was ≥ log2 1 . 5 and their adjusted p-value was ≤0 . 05 ., In the RNA seq analysis , it was observed that 437 and 425 genes were over-expressed in R20291::sinRR’ and in JIR8094::sinRR’ mutant strains , respectively , while 668 and 208 genes were under-expressed in R20291::sinRR’ and JIR8094::sinRR’ mutant strains , respectively ., Results from the transcriptome analysis confirm that as in B . subtilis , SinRR’ in C . difficile also regulates a wide range of genes involved in several pathways including sporulation , motility , metabolism , membrane transport , stress response and toxin synthesis ( Fig 2A ) ., A list of genes identified to be differentially regulated in mutants R20291::sinRR’ and JIR8094::sinRR’ compared to their parent strains are listed in S4 , S5 , S6 and S7 Tables respectively ., To test and validate the transcriptome profiles , we performed relevant phenotypic assays and functional analysis with parent and mutant strains for major pathways ( sporulation , motility , toxin production and autolysis ) that were suggested to be regulated by SinR and SinR’ ., We have included following strains in the phenotypic analysis: parent strain , sinRR’ mutant , sinRR’ mutant with pRGL311 ( plasmid with sinRR’ under its native promoter ) , and sinRR’ mutant with pRG334 ( plasmid with sinRR’ under the inducible promoter ) ., To determine the independent role of SinR and SinR’ in the phenotypes , the sinRR’ mutant with plasmids: pRG300 ( sinR gene alone with its promoter region ) ; pRG310 ( sinR under the inducible promoter ) ; and pRG306 ( sinR’ alone under the inducible promoter ) were used ., Western blot analysis with SinR and SinR’ specific antibodies were performed to confirm their expressions from the constructs , and the sinRR’ mutant with vector alone was used as negative controls ( Fig 2B ) ., Growth curve analysis showed when sinRR’ was expressed from its promoter or the inducible promoter in the sinRR’ mutant , no autolysis was observed , and they grew similar as the wild type ( Fig 2C and S4 Fig ) ., In the Triton X-100 autolysis assay , a partial recovery from autolysis was observed when either SinR or SinR’ alone was expressed in the mutant ( S4 Fig ) ., To determine the role of sinRR’ on sporulation , we grew the test strains on 70:30 sporulation agar for 30h ., Initial analysis through phase contrast microscopy detected no spores in R20291::sinRR’ ( Fig 3A ) ., Transmission electron microscopy ( TEM ) further confirmed this observation ( Fig 3B ) ., Fully mature spores could be detected in R20291 , whereas the sinRR’ mutant cells were devoid of any spores ., Similar results were obtained for JIR8094::sinRR’ mutant as well ( S5 Fig ) ., We performed ethanol treatment based sporulation efficiency assay where the ability of the bacteria to produce viable spores were analyzed by counting the total number of CFU ( Colony Forming Units ) following ethanol treatment ., The mean sporulation efficiency of the parental strain R20291 was 18 . 7% ( Fig 3C ) ., The sinRR’ mutant strain did not produce any spores , and the percentage of sporulation was near zero ., We were surprised by the observation that expression of either sinRR’ or sinR/sinR’ alone also did not revive the sporulation in the sinRR’ mutants ( Fig 3C ) ., Sporulation in C . difficile is initiated with the activation of Spo0A , which in turn triggers early sporulation gene transcription 22 , 40 ., Transcripts of spo0A were 3 . 5-fold and 2 . 9-fold under-expressed in JIR8094::sinRR’ and in R20291::sinRR’ strains respectively , when compared to parent strains ., We performed western blot analysis with the Spo0A specific antibodies 41 ., We detected GDH ( glutamate dehydrogenase ) for loading control since its production was found to be unaffected in the sinRR’ mutants ., Western blot analysis showed that in R20291::sinRR’ the Spo0A was absent or below the detectable level ( Fig 3C , S5 Fig ) ., Lower production of Spo0A can result in down-regulation of all sporulation genes under its control ., Our transcriptomic data indeed found many sporulation-associated genes to be affected ( Tables 1 , S4 and S6 ) in the sinRR’ mutant ., The QRT-PCR analysis performed on selected sporulation genes confirmed their down-regulation in the sinRR’ mutants ( S8 Table ) ., Since our transcriptome analysis and western blot analysis revealed a lower Spo0A in R20291::sinRR’ , we decided to test whether the asporogenic phenotype of the sinRR’ mutants is due to the lower production of Spo0A ., We expressed spo0A from its native promoter ( pRGL312 ) in the R20291::sinRR’ and production of Spo0A in sinRR’ mutants was verified through the western blot analysis using Spo0A specific antibodies ( Fig 3C ) 41 ., To our surprise , production of Spo0A in the sinRR’ mutants did not induce the sporulation in the R20291::sinRR’ strain ( Fig 3C ) ., For sporulation to proceed normally , the Spo0A protein should get activated by phosphorylation 42 ., Spo0A~P then acts as a transcriptional activator for many downstream genes in the sporulation pathway that includes sigma factors , the forespore specific sigF , and the mother cell-specific sigE 22 , 40 , 42 ., We performed QRT-PCR to detect the transcripts of Spo0A~P activated sigF and sigE genes ., We did not observe increases in sigF and sigE transcript levels in the spo0A expressing sinRR’ mutant when compared to the sinRR’ mutant with vector alone control ., This result suggests that activation of Spo0A to Spo0A~P is affected in the sinRR’ mutant ., In Bacillus sp ., , the pathway that controls Spo0A phosphorylation is well characterized 43–47 ., In Clostridia , the components of this phosphorelay are absent , and it has been hypothesized that sporulation-associated sensor kinases may directly phosphorylate the Spo0A for its activation ., In C . difficile , four orphan kinases ( CD630_01352 , CD630_2492 , CD630_01579 , and CD630_1949 ) are present , among which , the CD630_1579 kinase was shown to phosphorylate Spo0A in vitro , and the CD630_2492 mutant was found to be less efficient in sporulation 48 ., In the transcriptome data , the CD630_1579 and the CD630_ 2492 kinases were to be under-expressed ~1 . 5-fold and ~3-fold , respectively , in the JIR8094::sinRR’ mutant ., However , their homologs CDR20291_1476 and CDR20291_2385 in the R20291::sinRR’ were not affected suggesting that these kinases might not be the main reason for Spo0A inactivation in the sinRR’ mutants ., Since the regulatory network of Spo0A activation is largely unknown , there is a possibility that unknown kinases could have been affected in sinRR’ mutants ., The JIR8094 strain was intrinsically non-motile due to mutations within the flagellar operon 49 ., Hence , we choose only R20291 and R20291::sinRR’ to perform motility-related experiments ., The R20291::sigD mutant and the R20291::sinRR’ strains with vector alone ( pRPF185 ) were used as the controls ., Exponentially growing bacterial cultures were spotted on BHI with 0 . 3% agar and was incubated at 37°C for 36h to monitor motility ., The bacterial cultures expressing sinRR’ , or sinR or sinR’ from the tet-inducible promoters were spotted on BHI with 50 ng/ml of ATc and 0 . 3% agar ., In the motility assays , the R20291::sinRR’ strain was defective in motility ( Fig 4C and S6 Fig ) ., The transcriptome analysis supported our observation , where sigD , the sigma factor needed for the transcription of the flagellar operons , was found to be 14-fold under-expressed in the R20291::sinRR’ ( Fig 4A , S4 Table ) along with other motility-related genes ., Electron microscopic analysis followed by negative staining failed to detect flagellar structures in the R20291::sinRR’ ( Fig 4B ) ., A dot blot analysis with FliC ( the flagellar structural protein ) specific antibodies also confirmed the absence of flagella in the R20291::sinRR’ strain ( S6 Fig ) ., Expression of sinRR’ from its promoter or the inducible promoter revived the motility ( Fig 4C ) ., Interestingly , expression of SinR alone was sufficient to revive the motility in the R20291::sinRR’ strain , whereas the SinR’ expression alone did not have any effect ( Fig 4C ) ., SigD is needed for the transcription of the flagellar operon in C . difficile 25 , 26 ., To determine whether the non-motile phenotype of sinRR’ mutant is due to the reduced levels of sigD in the sinRR’ mutants , we expressed sigD from the tetracycline-inducible promoter by introducing the construct pRGL291 into the R20291::sinRR’ strain ( S1 ) ., We observed motility was partially restored in the R20291::sinRR’ when the sigD expression was induced ( Fig 4C ) , suggesting that sinRR’ controls motility by controlling the expression of sigD in C . difficile ., The transcriptome analysis and the follow-up QRT-PCR ( Fig 5A , Table 2 , S4 , S6 and S8 Tables ) result suggested sin locus’s role in toxin gene regulation ., Toxin ELISA was performed with the cytosolic protein extracts of sinRR’ mutants and their respective parent strains ., Bacterial cultures expressing either sinRR’ or sinR/sinR’ alone from the tetracycline-inducible promoter were grown for 6h in TY medium and were induced with 50ng/ml of ATc for 5 hours ., Cytosolic proteins harvested from these induced cultures were used for toxin ELISA ., We observed a six-fold reduction in toxin production ( Fig 5A ) in the R20291::sinRR’ when compared to the R20291 strain ., In JIR8094::sinRR’ however , a moderate two-fold reduction in toxin level was recorded when compared to the parent strain ( S7 Fig ) ., Expression of sinRR’ in the mutants brought the toxin production back to the level comparable to the parent strains ., As we observed in the motility assay , expression of sinR alone was sufficient to bring back the toxin production in the sinRR’ mutant , while expression of sinR’ did not show any effect ., In C . difficile , SigD positively regulates tcdR , the sigma factor needed for toxin gene transcription 25 , 26 ., Interestingly , the expression of sigD from an inducible promoter revived the toxin production in sinRR’ mutants , suggesting that sinRR’ controls both toxin production and motility by regulating sigD in C . difficile ., We observed that SigD expression in the sinRR’ mutants partially recovered both the motility and the toxin production in that strain ( Fig 4C and Fig 5A ) ., The main question that arises from this observation is how SinR controls sigD expression ., The sigD gene is part of the flagellar operon , whose transcription is directly controlled by the intracellular cyclic di-GMP ( c-di-GMP ) concentration 26 , 50 ., Within the cells , the c-di-GMP is synthesized from two molecules of GTP by diguanylate cyclases ( DGCs ) and is hydrolyzed by phosphodiesterases ( PDEs ) 50 , 51 ., The functionality of several of these C . difficile DGCs and PDEs has been confirmed by expressing them heterologously in Vibrio cholerae , where they resulted in phenotypes ( biofilm formation and motility ) that correspond to elevated or lowered levels of intracellular c-di-GMP 51 ., In C . difficile when CD630_1420 ( dccA ) was expressed from an inducible promoter , it resulted in elevated levels of intracellular c-di-GMP and reduced bacterial motility 50 ., In R20291::sinRR’ , ten-fold more ( -3 . 3 Log2 fold ) dccA ( CDR2029_1267 ) transcript was observed ( S5 Table ) compared to parent ., We measured the intracellular concentration of c-di-GMP ( S8 Fig ) and observed a nearly three-fold increase in the c-di-GMP concentration in the sinRR’ mutant compared to the parent R20291 strain ( Fig 5B ) ., This elevated intracellular level of c-di-GMP in sinRR’ mutants can block the sigD expression , which in turn will result in reduced motility and toxin production ( Figs 4C and 5B ) ., Hence , when sigD was expressed from the tetracycline-inducible promoter ( which is not affected by c-di-GMP concentration ) , motility and toxin production in the sinRR’ mutant could be revived ., These two findings corroborate our conclusion that elevated levels of c-di-GMP in sinRR’ mutant plays a major role in controlling its toxin production and motility ., We are currently performing experiments to test whether SinR can directly regulate dccA in C . difficile ., Results from the sinR and sinR’ complementation experiments showed that expression of SinR alone could revive the toxin production and the motility in the R20291::sinRR’ strain , whereas SinR’ expression alone did not have any effect on the toxin production or the motility ( Figs 4C and 5A ) ., These results suggested that among SinR and SinR’ , only SinR can directly influence the toxin production and the motility , which raised the question on the role of SinR’ in these pathways ., To find the answer , we created a sinR’ mutant which expressed SinR in the absence of SinR’ ( S9 Fig ) ., Our repeated attempts to create a sinR’ mutant using the similar technique in the JIR8094 background failed for unknown reasons ., Mutation in sinR’ was confirmed by PCR ( S9 Fig ) and western blot analysis using SinR’ specific antibodies ., As expected the SinR’ mutant produced SinR protein , but not the SinR’ ( S9 Fig ) ., The R20291::sinR’ grew almost similar to the parent strain and did not show any profound autolysis phenotype as the R20291::sinRR’ ( S6 Fig ) ., We performed the assays to measure sporulation , motility and toxin production in the R20291::sinR’ ., In the sporulation assay , it was found that R20291::sinR’ produced nearly three-fold more spores than the parent R20291 strain ( Fig 6A ) ., The R20291::sinR’ was more motile than the R20291 strain ( Fig 6B ) ., Similarly , a 2 . 5-fold increase in the toxin production was observed in the R20291::sinR’ when compared to the parent strain ( Fig 6C ) ., These initial results revealed that SinR’ can negatively influence sporulation , toxin production , and motility ., In our complementation of R20291::sinRR’ we showed that presence of SinR’ alone in the C . difficile cells in the absence of SinR could not influence either toxin production or the motility ( Fig 4C and Fig 5A ) ., Hence , SinR’ must be influencing these pathways through its action on SinR ., For example , if SinR’ is an inhibitor of SinR then the absence of SinR’ in the R20291::sinR’ would result in increased SinR activity , which in turn may result in increased sporulation , toxin production and motility in this strain ., To test this hypothesis , we performed two experiments ., First , tested the effect of over-expressed SinR in the wild-type strain; Second , we checked for physical interaction of SinR with SinR’ proteins by performing pull-down experiments ., The plasmid construct with either sinR ( pRG300 ) or sinR’ ( pRG306 ) under tetracycline-inducible promoter were introduced into R20291 parent strain and were tested for their toxin production , sporulation , and motility upon induction with ATc ., The R20291 strain with the vector alone was used as the control in these assays ., To perform the sporulation assay , we used bacterial cultures grown in 70:30 medium supplemented with 50 ng/ml of ATc for 36 hours ., Sporulation efficiency was enumerated as described in the method section ., Overexpression of sinR in R20291 strain increased its sporulation efficiency 2 . 5-fold ( 45% ) when compared to the control strain , where the average sporulation efficiency was 18% ., Overproduction of SinR’ in R20291 , however , reduced the sporulation efficiency to 5% ( Fig 7A ) ., Overproduction of SinR in R20291 resulted in increased motility as well ( Fig 7B ) ., In C . difficile , toxin production is minimal during exponential phase ( ~4 to 8h ) of the bacterial culture and reaches its maximum during the stationary phase ( 12h -16h ) 9 ., To detect any positive influence of both SinR and SinR’ on toxin production in the parent strain , we chose to use the 8h time point ., The bacterial cultures were grown for 6h in TY medium and were induced with 50 ng/ml of ATc for two hours before harvesting their cytosolic protein for Toxin ELISA ., Results from these experiments showed that overexpression of sinR resulted in a nearly 2 . 5-fold increase in the toxin production in the R20291 strain when compared to the R20291 with vector alone control ( Fig 7C ) ., No significant effect on toxin production was observed when sinR’ was overexpressed in R20291 ( Fig 7C ) ., This could be because sin locus is expressed only during the early stationary phase ( 10-12h ) in C . difficile ( S2 Fig ) ., We performed toxin ELISA at 8h time-point when SinR is predicted to be lower in the bacterial cells ., If SinR’ acts on toxin production primarily by repressing SinR , then overexpression of SinR’ at this time-point will not have any effect on toxin production ., Nevertheless , results from this overexpression studies demonstrated that increased SinR content in C . difficile could result in increased toxin production , motility , and sporulation ., In B . subtilis , SinR monomers bind with each other to form a homotetramer , which would then bind to upstream sequences of the target genes to repress their expression 34 , 52 ., SinI in B . subtilis binds with SinR and prevents the SinR homotetramer formation and thus blocks its activity 52 ., To test the protein-protein interaction of C . difficile SinR with SinR’ , we performed GST pull-down experiments using SinR-6His and SinR’-GST ., Purified SinR-6His protein was mixed with crude lysates from E . coli expressing SinR’-GST ., When we passed this mixture through the Ni++ affinity chromatography column , we pulled out SinR-6His along with SinR’-GST , suggesting the tight association of SinR with SinR’ ( Fig 8A , lanes 5 , 7 ) ., In control , the GST alone did not interact with the SinR-6His ( Fig 8A , lanes 6 , 8 ) , confirming protein specific interaction between SinR with SinR’ ., These results provided compelling evidence that SinR’ affects toxin production and sporulation indirectly by binding with SinR to inhibit its activity on its target genes ., Transcriptome analysis of the R20291::sinRR’ showed up-regulation of codY , an important global regulator by ~3 to 30 fold compared to parent strains ( S5 Table , S8 Table ) ., CodY is highly conserved in many Gram-positive bacteria 53–55 ., In B . subtilis it regulates several metabolic genes and controls competence , sporulation , and motility 56–58 ., In C . difficile , the codY mutant produced more toxins and spores than the parent strains and thus it is a repressor of these pathways 14 , 17 , 18 ., We hypothesized that many phenotypes and transcriptional changes we observe in the sinRR’ mutant could be related to the up-regulation of codY in these mutant strains ., To investigate whether SinR and SinR’ or both controls codY expression by binding to the promoter region of codY , we carried out electrophoretic mobility shift assays ( EMSAs ) ., We used radiolabeled DNA probe that contained the putative promoter region of the codY gene and performed binding reactions using purified SinR-6His or SinR’-6His proteins ., First , we tested SinR alone at increasing concentrations and found that it can shift the probe when used above 100 nM concentration ( Fig 7B ) ., When SinR’ was used similarly , it was unable to cause the mobility shift of the probe , even at the highest concentration ( Fig 7B ) ., We then tested whether SinR’ would prevent SinR from binding to the codY promoter region ., To do this , we used increasing amounts of SinR’ , in the presence of a fixed amount of SinR ( Fig 7B ) ., The results show that the presence of SinR’ in the reaction mix could prevent SinR from binding to the DNA ., As a negative control , we used a DNA probe that contained the promoter region of gluD , which codes for glutamate dehydrogenase ( GDH ) ., Neither SinR nor SinR’ was able to shift the control DNA even at the highest concentrations tested ( S10 Fig ) ., Based on these results , we conclude that SinR binds specifically to codY promoter region to control its transcription ., This result also provided evidence that the SinR’ interaction with SinR prevents its regulatory activity on its target gene ., In a recent study , CodY was found to negatively regulate sinRR’ expression in the C . difficile 630Δerm strain 14 ., A CodY putative binding site was identified in the sin locus upstream sequence , and reporter fusions with the sin locus promoter revealed the CodY could negatively regulate sin locus expression in this strain ., However , in the UK1 strain ( belongs to the ribotype 027 as R20291 ) , the promoter fusion revealed a positive regulation of sin locus by CodY ., Because of these contradictory observations , one could not conclude whether CodY regulates sin locus ., To examine the role of CodY on sin locus expression , we performed EMSA with purified CodY-6His and the putative CodY binding region upstream of sin locus ., An oligonucleotide with putative CodY binding sequence upstream of sinR was synthesized ( ORG 721 ) ( S2 Table ) and was radioactively labeled with γ- 32 P dATP ., A double-stranded DNA probe was generated after annealing with the complementary oligonucleotide ( ORG722 ) ., It is worth noting no sequence difference was found within this putative sin promoter regions

Introduction, Results, Discussion, Materials and methods

Clostridium difficile is the primary cause of nosocomial diarrhea and pseudomembranous colitis ., It produces dormant spores , which serve as an infectious vehicle responsible for transmission of the disease and persistence of the organism in the environment ., In Bacillus subtilis , the sin locus coding SinR ( 113 aa ) and SinI ( 57 aa ) is responsible for sporulation inhibition ., In B . subtilis , SinR mainly acts as a repressor of its target genes to control sporulation , biofilm formation , and autolysis ., SinI is an inhibitor of SinR , so their interaction determines whether SinR can inhibit its target gene expression ., The C . difficile genome carries two sinR homologs in the operon that we named sinR and sinR’ , coding for SinR ( 112 aa ) and SinR’ ( 105 aa ) , respectively ., In this study , we constructed and characterized sin locus mutants in two different C . difficile strains R20291 and JIR8094 , to decipher the locus’s role in C . difficile physiology ., Transcriptome analysis of the sinRR’ mutants revealed their pleiotropic roles in controlling several pathways including sporulation , toxin production , and motility in C . difficile ., Through various genetic and biochemical experiments , we have shown that SinR can regulate transcription of key regulators in these pathways , which includes sigD , spo0A , and codY ., We have found that SinR’ acts as an antagonist to SinR by blocking its repressor activity ., Using a hamster model , we have also demonstrated that the sin locus is needed for successful C . difficile infection ., This study reveals the sin locus as a central link that connects the gene regulatory networks of sporulation , toxin production , and motility; three key pathways that are important for C . difficile pathogenesis .

In Bacillus subtilis , sporulation , competence and biofilm formation are regulated by a pleiotropic regulator called SinR ., Two sinR homologs are present in C . difficile genome as an operon and henceforth labeled as sinR and sinR’ ., Our detailed investigation revealed that in C . difficile , the SinR and SinR’ are key master regulators needed for the regulation of several pathways including sporulation , toxin production , and motility .

bacteriology, medicine and health sciences, gut bacteria, toxins, pathology and laboratory medicine, gene regulation, pathogens, bacillus, microbiology, vertebrates, animals, mammals, toxicology, toxic agents, bacterial sporulation, prokaryotic models, experimental organism systems, bacteria, bacterial pathogens, research and analysis methods, clostridium difficile, hamsters, microbial physiology, medical microbiology, gene expression, microbial pathogens, genetic loci, pathogen motility, bacterial physiology, rodents, eukaryota, virulence factors, bacillus subtilis, genetics, biology and life sciences, amniotes, organisms

journal.pbio.1000027

Cryptic Variation in the Human Mutation Rate

The mutation rate is thought to vary across the human genome on several different scales ., At the chromosomal level , the Y chromosome evolves faster than the autosomes , which evolve faster than the X chromosome 1 , 2 ., This is thought to be due to males having a higher mutation rate than females ., The autosomes also appear to differ in their rates of mutation for reasons that are unclear 3 , 4 ., At the next level down , there appears to be variation in the mutation rate over a scale of several hundred kilobases 4 , 5 , another pattern that remains unexplained ., However , the most dramatic variation in the mutation rate is observed over fine scales in which adjacent sites can have very different mutation rates ., In the nuclear genome , this variation has been shown to be associated with context , the best-known example being the CpG dinucleotide in mammals ., CpG dinucleotides are generally methylated in mammals and since methyl-cytosine is unstable , this leads to a high rate of C→T and G→A transitions at these sites , which is about 10- to 20-fold higher than at other sites 6 , 7 ., However , the CpG effect is not the only source of fine-scale variation in the mutation rate; the rate of mutation appears to vary by about 2- or 3-fold as a function of other adjacent nucleotides 8–11 ., Although variation in the mutation rate has been well-characterised in terms of adjacent nucleotides 8 , 9 , 11 , it is possible that there is other variation in the mutation that is associated with either distant or complex context effects , which has hitherto escaped detection ., We investigated this question by testing whether human and chimpanzee single nucleotide polymorphisms ( SNPs ) occur at orthologous sites in the genome ., If there is variation in the mutation rate , we expect to see an excess of sites at which both humans and chimpanzees have a SNP ., To investigate whether human and chimpanzee SNPs tend to occur at the same sites in the genome , we BLASTed all chimpanzee SNPs against a dataset of human SNPs ., This yielded a dataset of 309 , 158 alignments of 81 base pairs ( bp ) with the chimpanzee SNP in the central position and a human SNP elsewhere within the alignment ., Of these alignments , 11 , 571 have the human and chimpanzee SNP at the same position ( Figure 1 ) ; we refer to these as coincident SNPs ., This number of coincident SNPs is much greater than the 3 , 817 we would expect if the human SNPs were distributed at random across the alignment , and also much greater than the 6 , 592 we would expect taking into account the influence of the adjacent nucleotides on the mutation rate , henceforth known as “simple” context effects ., The observed excess of coincident SNPs is significantly greater than the expected number ( ratio of observed over expected with simple context effects = 1 . 76 , with a standard error of 0 . 02 , p < 0 . 0001 under the null hypothesis that the ratio is 1 ) ., This excess is not due to our inability to correct for CpG effects; if we remove CpG dinucleotides from the analysis , we observe 5 , 028 coincident SNPs but would only expect 2 , 533 taking into account simple context effects ( ratio = 1 . 98 ( 0 . 03 ) ; p < 0 . 0001 ) ., If we look at the pattern of coincident SNPs , it is evident that almost all the excess is due to the same SNP being present in both humans and chimpanzees , with A-T/A-T SNPs being dramatically over-represented ( Table 1; see Table S1 for the analysis with CpG sites removed ) ., Although the excess of coincident SNPs is consistent with variation in the mutation rate that is not associated with simple context , there are several other explanations that warrant consideration ., In correcting for simple context effects , we have also made two assumptions; we have assumed that the pattern of mutation is the same on the two strands of the DNA duplex , and we have assumed that context effects are the same across the genome ., As a consequence of these assumptions , we could be underestimating the expected number of coincident SNPs ., For example , let us imagine that the triplet AAA has a high mutation rate on one strand , say the transcribed strand , and a low mutation rate on the other strand , but that the pattern is the opposite for the triplet CCC ( note that when we refer to the mutation of a triplet , we are referring to the mutation rate of the central nucleotide ) ., Because the relative mutation rates of AAA and CCC depend on which strand we are considering , we would tend to underestimate the expected number of coincident SNPs ., The pattern of mutation is known to differ between the two DNA strands in a manner that depends on transcription 12 , 13 ., However , what is important for our analysis is whether the relative mutation rates of the triplets differ between strands; it is the relative , rather than the absolute rate , that matters , because for each alignment we calculate the chance of a coincident SNP relative to the chance that the human SNP occurs at one of the other triplets in the sequence ., To investigate this , we estimated the mutation rate of the central nucleotide in each triplet for a set of human genes for which we knew the direction of transcription; we also considered a subset of these genes known to be expressed in the testis ., In agreement with Green et al . 12 , we observe a 25% excess of A→G transitions over T→C transitions; however , we did not observe an excess of G→A transitions over C→T transitions , even in our testis-expressed genes ., Crucially for our analysis , the mutation rate of each triplet is highly correlated to its reverse-compliment triplet for all genes ( Pearson correlation coefficient r = 1 . 00 for all triplets , r = 0 . 85 without triplets containing CpGs; Figure S2A ) and for genes expressed in the testes ( r = 0 . 99 for all triplets , r = 0 . 75 without triplets containing CpGs; Figure S2B ) ; genes expressed in the testes are expressed in the male germ-line , where any strand asymmetry in the pattern of mutation will have an evolutionary effect ., It therefore seems unlikely that strand asymmetry in the pattern of mutation is leading to an underestimate of the expected number of coincident SNPs ., The excess of coincident SNPs could also be due to variation in the pattern of mutation across the genome for reasons similar to those given for strand asymmetry; if the relative rate at which each triplet mutates differs between genomic regions , then we will underestimate the expected number of coincident SNPs ., Since such variation in the pattern of mutation might be expected to generate differences in base composition , we divided our dataset of alignments according to their GC content and estimated the mutation rate of the central nucleotide in each triplet in the chimpanzee sequence using the human sequence to infer the ancestral sequence ., The relative rates of mutation inferred from the sequences in the upper and low GC content quartiles are highly correlated to each other ( r = 0 . 99 using all triplets; r = 0 . 88 excluding triplets involving CpGs; Figure S3 ) , which suggests that triplets that are highly mutable in high–GC content sequences also tend to be highly mutable in the low–GC content sequences ., It therefore seems unlikely that we are underestimating the expected number of coincident SNPs because of variation in the pattern of mutation ., As expected , we find a significant excess of coincident SNPs in both the upper and lower GC quartile datasets , although the excess of coincident SNPs appears to be slightly stronger in GC-poor DNA ( Table S2 ) ., The excess of coincident SNPs could be due to inheritance , in humans and chimpanzees , of polymorphisms that were present in their last common ancestor ., Two lines of evidence suggest that this is not the case ., First , we repeated the analysis using human and macaque SNPs ., Since these two species diverged more than 23–34 million years ago ( Mya ) 14 , as opposed to the 6–10 My that separates human and chimp 14 , one would expect very few polymorphisms to be shared between human and macaque ., However , in this dataset we also see a significant excess of coincident SNPs whether we consider all sites ( ratio = 1 . 64 ( 0 . 19 ) ; p < 0 . 001 ) or non-CpG sites ( 1 . 51 ( 0 . 26 ) ; and p < 0 . 05 ) ., Second , the pattern of coincident SNPs ( Table 1 ) is inconsistent with ancestral polymorphism ., All four of the possible transversion SNPs are approximately equally common amongst SNPs in general ( proportion of transversions amongst human SNPs: G/T = 0 . 092 , C/A = 0 . 091 , C/G = 0 . 088 , A/T = 0 . 075; transitions: C/T = 0 . 33 , G/A = 0 . 33 ) ., We would therefore expect a G-C SNP in chimps to be coincident with a G-C SNP in humans approximately equally often as an A-T SNP in humans is coincident with an A-T SNP in chimps ., However , we see distinct biases , with coincident A-T/A-T SNPs being much more common than the other transversions ., It is also possible for the apparent excess of coincident SNPs to be due to selection; if some regions of the genome are under selection , then we expect them to have a low density of SNPs , because many SNPs will be removed as they are deleterious ., As a consequence , SNPs will be clustered between these regions , causing an apparent excess of coincident SNPs ., This seems an unlikely explanation , since the vast majority of our data is intergenic and intronic ( 98% and 99% of the human and chimpanzee SNPs in our BLAST databases , respectively ) , and although selection is known to act within these regions , it is thought to only affect a small percentage of sites 15–17 ., Furthermore , if selection was causing an excess of coincident SNPs , we would expect SNPs to be clustered generally , but this is not observed ( Figure 1 and Figure S1 ) ., There is a small excess of human SNPs adjacent to the chimpanzee SNP , but this is a consequence of CpG effects—the chimpanzee SNP is disproportionately likely to occur within a CpG , which means that a human SNP is also likely to occur at the same site , or at an adjacent site ., If we remove CpGs , this slight excess of adjacent SNPs disappears ( Figure S1 ) ., Otherwise there is no tendency for SNPs to cluster ., It therefore seems that the excess of coincident SNPs is a consequence of variation in the mutation rate that is not associated with simple context effects , variation in these context effects between strands or regions of the genome , or natural selection ., The question therefore arises whether the variation in the mutation rate is associated with other contexts that are distant from the target site , degenerate in nature , or sufficiently complex to be difficult to discern ., It should be noted that simple context effects beyond the adjacent nucleotides ( e . g . , 1 bp removed from the target site ) are not responsible for the excess ., Although these effects exist 11 , they are much smaller than those of adjacent nucleotides , which themselves have a relatively modest effect if we remove CpGs; e . g . , the expected number of non–CpG coincident SNPs is 2 , 115 if we ignore adjacent nucleotide effects , and it is 2 , 533 if we include these effects ., To investigate whether there are other , more complex context effects , we tabulated the frequency of each triplet at each site in the alignments containing coincident SNPs , and a similar-sized dataset of alignments with noncoincident SNPs ., Surprisingly , we found significant heterogeneity in triplet frequencies that extends to about 80 bp on either side of the coincident SNP ( Figure 2A ) ; i . e . , the relative frequencies of the triplets at sites close to the coincident SNP are different from the average across the alignments ., In contrast , if we consider alignments without a coincident SNP , but with a chimpanzee SNP , we only see significant heterogeneity in triplet frequencies within 10 bp of either side of the SNP ( Figure 2B ) ., Despite the heterogeneity in triplet frequencies surrounding a coincident SNP , we could discern very few patterns in the triplets that are over- or under-represented ., The only conspicuous pattern is an excess of TTT triplets upstream and AAA triplets downstream of coincident SNPs ., However this seems to explain little of the overall excess of coincident SNPs ., If we repeat the analysis but remove all cases in which there is a run of three or more nucleotides , of any type , with or without SNPs within them , then from our alignments we find 8 , 536 alignments with a coincident SNP versus an expected number of 4 , 434 , taking into account simple context effects ( ratio = 1 . 93 ( 0 . 02 ) ; p < 0 . 0001 ) ., Considering pentamers , rather than triplets , also fails to reveal any context that is associated with coincident SNPs , except for the α-polymerase pause site motif , TG ( A/G ) ( A/G ) ( G/T ) ( A/C ) , which has been suggested as a hypermutable motif 18 , 19 ., However , we only observe an excess of α-polymerase pause sites immediately downstream of coincident SNPs , and the total number of coincident SNPs explained by this motif is trivial ( 2 . 2% ) ., To quantify the level of cryptic variation in the mutation rate , we fit two models to the ratio of the observed number of coincident SNPs over the number expected with simple context effects ., In the first model , we assumed that the variation in the mutation rate was log-normally distributed; in the second , we assumed that there were two types of sites—normal and hypermutable ., These models give qualitatively similar estimates of the variation , so we only discuss the log-normal model in detail , because this is a model with a single parameter ( details of the two-rate model are given in Text S1 ) ., Because our method for controlling for simple context effects tends to underestimate the expected number of coincident SNPs when we have CpG sites , we concentrate on non-CpG sites ., We fit two sub-models to our data ., In the first , we assume that the mutation rate of a site is invariant in both humans and chimpanzees ., Under this “static” model , we estimate the shape parameter of the log-normal to be 0 . 83 ( 95% confidence intervals ( CIs ) of 0 . 81 , 0 . 84 ) for non-CpG sites ., However , this model may not be realistic , since we might expect sites with high mutation rates to destroy themselves; e . g . , if a site has a high rate of C→T mutation , then it will rapidly become fixed for T and therefore become nonhypermutable ., We therefore also fit a model in which the time a site remains at a certain mutation rate depends upon that mutation rate , assuming an average divergence between humans and chimpanzees of 0 . 92% for non-CpG sites 20 ., Under this model , we estimate slightly higher levels of cryptic variation: we estimated the shape parameter to be 0 . 85 ( 0 . 83 , 0 . 87 ) —higher shape parameters mean more variation ., The level of variation that these distributions represent is considerable; with a shape parameter of 0 . 85 the fastest 5% of sites mutate at least 16 . 4-fold faster than the slowest 5% of sites ., This level of variation in the mutation rate is greater than the variation associated with simple context: the variance due to simple context , including CpGs , is 0 . 59 , whereas the variance due to cryptic variation at non-CpG sites is 1 . 05 ., However , this large difference in variance might be due to the model ., If we consider a simple two-rate model in which sites are either hypermutable or normal , and constrain the proportion of hypermutable sites to be 2% , which is the proportion of sites that are involved in CpGs in the human genome 21 , then we estimate that hypermutable sites would have to mutate 9 . 3-fold faster than normal sites to explain the excess of coincident SNPs ., This is similar to 10–20-fold higher rate that CpGs mutate 9 , 20 ., We have shown that there is an excess of sites that have a SNP in both the human and chimpanzee genomes ., We demonstrated that this is not due to neighbouring nucleotide effects , shared ancestral polymorphism , or natural selection ., It therefore seems that this excess is due to variation in the mutation rate that is not associated with simple context effects and is cryptic in nature ., We also show that triplet frequencies surrounding sites with coincident SNPs are highly nonrandom , but we have been unable to discern any specific motifs in these regions ., This suggests that there are probably complex context effects that extend some distance from the site they effect ., Furthermore , we show that there has to be considerable variation in the mutation rate to explain the observed excess of coincident SNPs ., The presence of such cryptic variation in the mutation rate is perhaps not surprising given the evidence that some sites in the human mitochondrial genome are hypermutable ., Hypermutation had long been suspected based on the excess of homoplasies in human mitochondrial DNA ( mtDNA ) phylogenies ( e . g . , see 22 ) and although such an excess could be due to hypermutation or recombination 23 , two recent analyses have provided convincing evidence that the excess is due to hypermutation ., Stoneking 24 showed that mitochondrial mutations in human pedigrees tend to occur at sites that have high levels of homoplasy , and Galtier et al . 25 have recently shown that synonymous mitochondrial SNPs tend to occur at the same positions in different species ., However , although many of the hot spots in mtDNA appear to be due to strand slippage–type mutational mechanisms 26 , 27 , this does not appear to be case for the cryptic variation in the mutation rate in nuclear DNA that we describe here ., There are two slippage mechanisms that can operate: template strand and primer strand dislocation ., Template strand dislocation is controlled for in our simple context analysis , and primer strand dislocation is controlled for in the analysis of homonucleotide runs ., It has also been shown recently that the mutation rate is elevated close to insertion and deletion mutations in the nuclear genomes of several eukaryotes , including humans 28 ., However , it seems unlikely that this process is generating the excess of coincident SNPs ., Indels appear to increase the rate of mutation but not at specific sites; rather the mutation rate is elevated close to an indel and this elevation in the mutation rate declines over several hundred nucleotides ., This would manifest itself as general tendency for SNPs to cluster , which we do not observe ( Figure 1 and Figure S1 ) ; we only observe a large excess of coincident SNPs and a small excess of adjacent SNPs ., Furthermore , humans and chimpanzees would both have to have segregating indels in the same locality to generate an excess of coincident SNPs ., Over the last few years , DNA sequence analysis has revealed that the mutation process is highly complex , varying between different parts of the genome and between different sites ., Unfortunately we do not yet understand many of these patterns ., We downloaded human and chimpanzee SNPs from dbSNP build 126 ., Dividing the data into chromosomes , we BLASTed each chimpanzee SNP , along with 50 bp of flanking DNA on either side of the SNP , against a database of human SNPs ., We set the BLAST parameters as follows; e-value = 1 × 10−30 , mismatch score = −1 , and simple sequence filter off ., We retained those alignments , which were 101 bp in length , and in which the human or chimpanzee sequence showed identity at 96 sites if the SNPs were coincident , or 94 sites if they were not coincident ., We adjusted the number of matches required to control for the fact that if the SNPs are not coincident , then there must be two extra mismatches ., We randomly chose one alignment if a chimpanzee SNP matched more than one human SNP at the levels of identity we set; we obtained very similar results removing these cases from the analysis ., The alignments were trimmed to 40 bp on either side of the central chimpanzee SNP because there is a slight bias away from finding human SNPs at the edges of the chimpanzee query sequence ., This bias occurs because SNPs , being classed as mismatches , tend to cause BLAST to prematurely terminate the alignment ., To perform the analysis of triplet frequencies , we downloaded an extended flanking sequence for the chimpanzee SNPs analysed ., The macaque SNPs were kindly provided by Dr . Ripan Malhi 29 ., We repeated the analysis as we did for chimpanzee but we relaxed the criteria used to identify orthologous human sequences containing SNPs to 86 matches if there was a coincident SNP , and 84 if there was not , with the e-value adjusted to allow this level of similarity to be found ., Sites were designated as CpG if the site , or any of the SNPs at the site , would yield a CpG dinucleotide ., We estimated the expected number of coincident SNPs , taking into account the effects of adjacent nucleotides on the rate of mutation , what we term “simple” context effects , as follows ., Our data consist of a set of alignments in which we have both a human and a chimpanzee SNP ., We start by tabulating the numbers of each triplet , nxyz , where x , y , and z can be T , C , A , or G , in the chimpanzee sequence in the alignments , along with the number of chimp triplets that have a human SNP opposite the central nucleotide , nxyz . Hsnp ., From these , we can estimate the probability of observing a human SNP opposite a chimpanzee triplet in our alignments: pxyz = nxyz . Hsnp / nxyz ., We can also calculate the frequency of each triplet in the chimpanzee sequences: fxyz = nxyz/Σnxyz To calculate the probability that the human and chimpanzee SNPs are coincident , we need to take into account that there are two alleles in the chimpanzee SNPs , and the triplets they are a part of will have different probabilities of having a human SNP opposite them ., If we knew the relative frequencies of the chimpanzee alleles , we could calculate the chance of a coincident SNP as gy pxyz + ( 1– gy ) pxyz where y and y are the two chimpanzee alleles and gy is the frequency of the y allele ., However , we do not have allele frequency information , so we estimated the relative probabilities of each of the two ancestral states for the chimpanzee SNP , since the ancestral allele is likely to be at a higher frequency in the population ., For example , let us imagine we have a CYC SNP—i . e . , a Y SNP surrounded by C on both sides ., The ancestral triplet could have been CCC or CTC ., The probability that the SNP was generated from a CCC can be estimated as mCCC = fCCC rCCC/ ( fCCCrCCC + fCTCrCTC ) where rxyz is the rate at which triplet XYZ generates a SNP in the central position of the triplet ., We estimate rxyz by orienting the chimp SNPs using the human sequence , excluding coincident SNPs and SNPs for which the human nucleotide is different to both chimp alleles; let sxyz . Csnp be the number of chimp triplets that are inferred to have generated a SNP , then rxyz = sxyz , Csnp/nxyz ., The expected number of coincident SNPs in each alignment is then , using the above example , ( mCCCpCCC + mCTC pCTC ) /Σpxyz , where the summation is across all the triplets in the alignment ., The total number of expected coincident SNPs was simply the sum across alignments ., We used two methods to calculate the standard error for the ratio of the observed number of coincident SNPs over the expected number: we bootstrapped the data by alignment and then summed the observed and expected values across the bootstrapped datasets ., However , it turned out that this was very closely approximated by assuming that the observed number of coincident SNPs was Poisson distributed and the expected value was known with no error; these are the standard errors we present ., We performed a number of simulations to check that the BLAST analysis was not biased and that our method to estimate the number of coincident SNPs under simple context effects worked well ., In each simulation , we evolved human genomic sequences under a mutation pattern , in which the mutation rate depended on the adjacent nucleotides , to generate a simulated human and chimpanzee sequence ., Into these we introduced SNPs according to the same mutation pattern at the density found in dbSNP—one SNP every 266 bp in humans and every 2 , 128 bp in chimp ., We then constructed a BLAST database of ∼140 , 000 human SNPs with 100 bp of flanking DNA sequence , and a query dataset of ∼18 , 000 chimpanzee SNPs with 50 bp of flanking DNA ., We ran the BLAST analysis and analysed the output exactly as we had with the real data ., We ran simulations in which we had no mutation bias and datasets in which the mutation rate of all triplets was the same except for triplets containing CpGs , which had a mutation rate 10 , 15 , or 20 times the background rate ., We ran a set of simulations in which we had 0% , 1% , and 2% divergence ., Our method works well at all divergences and under all mutation patterns , except when the CpG rate is very high , where the method tends to underestimate the expected number of coincident SNPs ( Table S3 ) ., Surprisingly , the method tends to slightly overestimate the expected number of coincident SNPs when CpG sites are removed for reasons that are not clear ., To investigate strand asymmetry , we estimated the mutation rate of the central nucleotide in each triplet by tabulating the number of times each triplet contained a SNP ., The direction of mutation was inferred from the frequency; i . e . , the minority allele was judged to be the new mutation ., We inferred mutation rates across 964 human genes from the Seattle SNPs 30 and Environmental Genome Projects 31 ., To investigate which of these genes are expressed in the male germ line , we downloaded gene expression data from the human testis from the study of Ge et al . 32 ., We obtained raw CEL files of gene expression levels from the NCBI Gene Expression Omnibus database ( http://www . ncbi . nlm . nih . gov/projects/geo/ ) ., We normalized the results from the mouse and rat arrays separately using the RMA algorithm 33 as implemented in Bioconductor 34 ., We judged a gene to be expressed within the testis if its expression was above 200 35 ., We estimated the variation in the mutation rate as follows ., We start by assuming there is no divergence between humans and chimpanzees so a hypermutable site in humans will also be hypermutable in chimpanzees ., Let the average probability of detecting a SNP at a site in humans and chimpanzees be μh and μc , respectively; if μh and μc are small , the probability at a particular site will be γμh and γμc , where γ is the relative rate of mutation ., Let us assume that γ takes some distribution D ( γ ) which has a mean of one ., The expected number of coincident SNPs is If there is no variation in the mutation rate then this reduces to such that the ratio of the number of coincident SNPs , over the number expected with no variation , is an equation which only depends upon the distribution of γ ., We assume that γ is either log-normally distributed , or that it has a two state distribution in which sites can either be hypermutable or normal ( see Protocol S1 ) ., We estimate the parameters of the distribution of γ by considering the ratio of the observed number of SNPs over the number expected with simple context effects ( i . e . , the number expected without cryptic variation in the mutation rate ) ., This model is unrealistic , because we assume that a site does not change its mutation rate; however , hypermutable sites are more likely to change , and this may lead them to become nonhypermutable ., Under the log-normal model , we assume that once a site changes , its mutation rate is drawn randomly from the log-normal distribution ., Let v be the average rate of mutation per unit time in both humans and chimpanzees ., Consider a site , in the ancestor of humans and chimpanzees , that currently has a mutation rate vγ ., The probability that the site will remain unchanged along both the human and chimpanzee lineage is where t is the time since humans and chimpanzees diverged ., The probability that such a site will produce a coincident SNP is If the site changes in one of the lineages , then the mutation rates in the two lineages become independent of one another; since the mean of a product is the product of the means , when two random variables are independent , the probability of a coincident SNP at a site which has undergone at least one substitution is The expected number of SNPs with no variation in the mutation rate is still P0 , as given by Equation 2 , so we can write the ratio of the expected number of coincident SNPs with variation over the expected number without variation in the mutation rate as This equation depends on the compound parameter 2vt , which is the average divergence between humans and chimpanzees and the distribution of γ ., Since we set the average of the log-normal distribution to one , we need only find the shape parameter of the log-normal distribution ., To estimate the variance associated with simple context effects , we calculated the mutation rate of each triplet as above , when correcting simple context effects ., We then scaled the mutation rates so the mean across triplets , taking into account their frequencies in the genome , had a mean of one ., We then calculated the variance ., This can be compared directly to the variance of the log-normal distribution which we had also constrained to have a mean of one ., We weighted the variance estimates from the CpG and non-CpG sites by the relative frequency of the sites .

Introduction, Results, Discussion, Materials and Methods

The mutation rate is known to vary between adjacent sites within the human genome as a consequence of context , the most well-studied example being the influence of CpG dinucelotides ., We investigated whether there is additional variation by testing whether there is an excess of sites at which both humans and chimpanzees have a single-nucleotide polymorphism ( SNP ) ., We found a highly significant excess of such sites , and we demonstrated that this excess is not due to neighbouring nucleotide effects , ancestral polymorphism , or natural selection ., We therefore infer that there is cryptic variation in the mutation rate ., However , although this variation in the mutation rate is not associated with the adjacent nucleotides , we show that there are highly nonrandom patterns of nucleotides that extend ∼80 base pairs on either side of sites with coincident SNPs , suggesting that there are extensive and complex context effects ., Finally , we estimate the level of variation needed to produce the excess of coincident SNPs and show that there is a similar , or higher , level of variation in the mutation rate associated with this cryptic process than there is associated with adjacent nucleotides , including the CpG effect ., We conclude that there is substantial variation in the mutation that has , until now , been hidden from view .

Understanding the process of mutation is important , not only mechanistically , but also because it has implications for the analysis of sequence evolution and population genetic inference ., The mutation rate is known to differ between sites within the human genome ., The most dramatic example of this is when a C is followed by G; both the C and G nucleotides have a rate of mutation that is between 10- and 20-fold higher than the rate at other sites ., In addition , is it known that the mutation rate may be influenced by the nucleotides flanking the site ., Here we show that there is also very substantial variation in the mutation rate that is not associated with the flanking nucleotides , or the CpG effect ., Although this variation does not depend upon the adjacent nucleotides , there are nonrandom patterns of nucleotides surrounding sites that appear to be hypermutable , suggesting there are complex context effects that influence the mutation rate .

genetics and genomics, evolutionary biology

Substantial variation is seen in the rate of mutation among different sites in the human genome, which is not associated with mutable CpG sites or due to simple context effects.

journal.pcbi.1006704

Protein—protein binding supersites

Specific protein-protein interactions are essential for maintaining a robust phenotype ., A deeper understanding of these interactions would allow the identification of cognate ligands1 and drivers of specificity , opening a pathway to manipulating the corresponding interaction interfaces in drug design applications2 ., While it has been estimated that a protein on average interacts with 3–10 other proteins3 , the Protein Data Bank4 ( PDB ) contains a disproportionally small fraction of known protein complexes ., For most of the PDB entries neither the ligand protein nor the protein binding interfaces are known ., In response to this important problem , a number of methods have been developed to predict protein binding interfaces using structural information , which may be available in the form of known experimental or computational three dimensional models5 ., The methods to predict protein interfaces can be grouped into two main approaches: ( 1 ) homology-based and ( 2 ) ab initio ., Homology-based predictions of interfaces rely on the knowledge of known protein complexes to infer the likely binding sites in similar proteins ., These methods can be very powerful6 , 7 , but their applicability is limited by the amount of known interfaces ., Within the category of “ab initio” protein interface predictions a number of studies have attempted to identify distinctive features of interfaces8–14 often employing various machine learning approaches ., These features include residue composition15 , residue conservation16–18 , hydrophobicity19 , 20 , planarity15 , predicted secondary structural features14 , 21 , electrostatics22 , accessible surface area , among others ., Some studies found that different subtypes of protein interfaces ( e . g . transient interfaces , interfaces between homo- and heteromers , etc . ) have distinct sequence features , which can be exploited to predict some of the interface residues from sequence14 , 23 ., For example , these features suggest that interfaces for obligate complexes are somewhat more hydrophobic and larger than other interfaces15 , 24 ., Similarly , binding site hot-spots have been predicted using residue composition , conservation analysis , or other structural features such as desolvation effects13 ., However , a generic conclusion after many studies and using larger and more diverse test sets is that protein interfaces do not have a specific composition or other universal features they share18 , 25 , 26 ., This is arguably the expected conceptual conclusion as it is difficult to conceive a universal external evolutionary pressure that would unify interfaces27 ., Current success rates for protein binding interface predictions on a residue level are just barely statistically significant when compared to random predictions28 ., Relevant to the current study are the works that discuss the possible generality of binding site locations , both for small molecule and protein ligands ., In the case of the former , it has been observed as early as in the 1980s that small organic molecules , both substrates and non-substrates tend to bind to similar , energetically favored “sticky” sites irrespective of their relevance to the target ., These observations were made by experimental studies that soaked target proteins in organic solvents and examined the crystal29 or NMR30 structure for invariable small molecules sticking to energetically favorable sites ., Computational methods such as the GRID31 , or the Multicopy Simultaneous Search ( MCSS ) 32 approach , as well as some of the most competitive methods currently available33 , are also based broadly on this observation ., It was observed in the late 1990s that protein superfolds ( frequently occurring proteins that share their overall structural topology but have a range of distinct functions ) have “supersites” ., In other words , despite substantial sequence divergence and the evolved distinct functions , the 10–15 superfolds that dominate about half of the structural fold population of the genomes34 usually have very similar binding site locations35 ., This observation was subsequently revisited and expanded to remote homologs with insignificant sequence similarity to the cognate ligands for a range of different fold topologies36 , 37 ., Docking programs have been used successfully to predict partner-specific interface residues such as the Atomic Contact Frequency ( ACF ) 38 or the Residue Contact Frequency ( RCF ) method39 and others40 ., These approaches require the prior knowledge of the cognate ligand from other , indirect sources , such as high throughput screening methods ., In the current work , we explored the generality of the phenomenon of binding supersites ., We report the surprising observation that protein-protein interaction sites serve as generic protein binding sites ., Protein ligands , irrespective of their relevance to the receptor protein , tend to bind to the cognate protein interface ., This behavior does not depend on the docking program used , the range and type of protein ligand probes employed , or more technical conditions such as the size of the binding sites considered ., Based on this new observation we introduce a docking-based , ab initio method for binding site prediction that does not require prior knowledge of the cognate ligand ., Binding interfaces are determined by the frequency of a receptor residue interacting with a range of unrelated protein ligands in extensive docking simulations ., A conceptual insight brought to light by our work is that protein shapes evolved to allow a surprisingly small number of suitable surface patches for interactions that are apparently sampled by a wide range of possible ligands ., Alternatively , it may be that a variety of unique residue patterns that evolved for recognizing a specific cognate protein ligand also present an energetically relatively favorable site for non-cognate proteins ., We explored the hypothesis of whether protein-protein interaction sites also serve as generic binding sites for a range of non-cognate ligands , and as such , behave similarly to protein-small-molecule-binding sites30 , 32 , 41 , 42 ., This would qualitatively generalize the observations made about supersites in superfolds35 ., We explored the preferred binding sites for a set of unrelated ligands on a large set of receptor proteins ., Surprisingly , we found that unrelated ligands have a strong tendency to dock to the same general area of a receptor as its cognate ligand ., We illustrate this in Fig 1 , where three , topologically different ligands ( all beta– 2jjs . C; mixed alpha and beta– 3h33 . A; and a small protein fold with few secondary structures– 2v86 . A ) , sharing no detectable structural or sequential similarity to the cognate ligand , all have a strong tendency to dock to the cognate protein binding site on the receptor protein ( 1cnz—3-Isopropylmalate dehydrogenase from Salmonella typhirium ) ., We explored the overall phenomenon by docking 13 different ligand probes , six immunoglobulin folds and seven randomly picked small protein folds on a combined target dataset of 24143 , 44 proteins with structurally defined protein binding sites ., We ranked the residues in the receptor protein based on the RIF score ( see Methods ) ., The statistical significance of the agreement of the top ranked residues and the cognate binding site was assessed by using hypergeometric distribution to model the probability of correctly selecting an interface residue by chance ., Out of the 241 target proteins , in 157 ±2 cases ( or 65 . 2 ± 0 . 9% ) the binding site was docked by a variety of unrelated ligands in a statistically significant manner ., We evaluated the performance by randomly selecting 2000 models from the total set of 26000 docked models ( 13 X 2000 per ligand probe ) and calculated the average performance and the standard deviation ., We further broke down results by complex and database type ., Performance on the Docking Benchmark44 and NOX43 databases were 70 . 3% ± 1 . 1 and 61 . 1 ± 1 . 6 , respectively ., Furthermore , the NOX database contained a relatively well-balanced set of obligate ( 73 ) and non-obligate complexes ( 60 ) , and the results on these subsets were 68 . 7 ±1 . 9 and 51 . 8 ±2 . 2 , respectively ., We also evaluated the results using sensitivity/specificity ROC curves ( Fig 2 . ) and obtained an Area Under the Curve value of AUC = 0 . 79 for the combined set , while 0 . 83 and 0 . 77 for the Docking Benchmark and NOX databases , respectively ., All these suggest that the observation about protein binding supersites is a generic feature of proteins , with some fluctuation of specific success rates depending on the choice of test database ., We also explored how well the cognate ligands bind to and define the annotated functional site of the receptor proteins in comparison to unrelated ligands ., ( Fig 3 ), Interestingly , while the cognate ligands have a tendency to better recognize the interface , this tendency is statistically not significantly different from the results obtained for unrelated ligands ., We further subdivided our results as a function of different ligand probes and ligand sizes , while also exploring two alternative docking programs , ZDOCK and GRAMM , to examine the role that variations in the scoring functions play in detecting supersites ., We found little dependence on the type of probe used with either docking program ( Fig 3 ) ., The differences in results obtained using individual probes are mostly statistically insignificant ., The success rate for the NOX dataset depending on the ligand probes ranged between 54 . 1 to 65 . 4% with an average success rate of 60 . 1% ± 3 . 9% using ZDOCK , while the success rate ranged from 36 . 8% to 51 . 9% with an average of 44 . 7% ± 4 . 4% using GRAMM ., ZDOCK appears to yield slightly better results with the immunoglobulin superfamily probes , while GRAMM works better with the non-immunoglobulin set of ligand probes ., If we use a consensus prediction from all 13 ligand probes , the performances in the case of ZDOCK and GRAMM are 60 . 1% and 44 . 7% , respectively ., The better performance of ZDOCK suggests that the energy function may play a role in defining the “stickiness” of protein binding supersites ., ZDOCK45 uses a statistical pair potential with a limited set of amino acid residue types , while the GRAMM46 energy function is arguably more general using a step function that includes a classic repulsion term ., We compared the actual interface residues predicted by the two docking programs , ZDOCK and GRAMM ., Although the entire set of interface residues predicted by the two docking programs were not identical , for 40% and 79% of the 241 proteins in the data set , the two docking programs predicted more than 10 or more than 5 interface residues in common out of 15 , respectively ., To put these numbers in a statistical context: the expected number of residues that are common out of 15 residues between any two random draws , —in protein sizes 100 , 150 , 200 , 250 and 300 are: 2 . 27 , 1 . 58 , 1 . 16 , 0 . 82 , and 0 . 81 residues , respectively ., Consequently , the two programs have a strong tendency to locate binding sites similarly ., The corresponding p-values of observed common residues between ZDOCK and GRAMM are all significant at any protein size ., We explored an additional aspect of the potential impact of the employed energy function ., ZDOCK ranks the generated docked poses by their energy score , so we explored if there is a difference in performance between the top-scoring and bottom-scoring docked poses ., Indeed , this phenomenon can be observed once we plot the performance of the first and last 200 docked poses ( Fig 4 ) ., There is a weak but persistent tendency that energetically higher ranked poses are more useful in identifying binding sites ( Fig 4 ) ., These small differences disappear as the number of sampled conformations approach 200 and beyond ., The differences between the accuracy of ZDOCK and GRAMM and between top-ranked and bottom-ranked docked poses of ZDOCK suggest that a more accurate energy function will identify binding sites more accurately because the relative affinity of non-cognate ligands will be better captured ., When considering the possible reasons for the existence of protein binding supersites , besides the general energetic preferences of certain “sticky” areas of the protein , one could also consider receptor-shape-driven causes ., For instance , one could speculate that in the case of small proteins it might be a geometrical artifact that only a confined area is suitable to accept interactions ., However , the distribution of the size of receptors in the current work has a large range ( <100 residues to >700 residues ) for which the ability to detect supersites appears to be uniformly high ( Fig 5 ) ., We further dissected the possible differences in performance between the two docking approaches ., First , we compared the performance of these techniques using 2000 models generated by the methods , irrespective of the size of the identified binding interface , with the performance when using only a subset of the docked complexes that have the most common interface sizes; in the current work , formed by 9 residues ( Table 1 ) ., Though the GRAMM docking method appears to sample a larger fraction of all the residues in the protein ( 85 . 9% vs 72 . 1% ) as well as the interface residues ( 99 . 6% vs 97 . 5% ) , ZDOCK identifies a larger number of true interface residues ranking in the top 15 positions ( 60 . 1% +/- 3 . 9 for ZDOCK vs . 44 . 7% +/- 4 . 4 for GRAMM ) ., In case of considering 9–residue patches only , as expected , the total number of residues sampled ( 40 . 7% for ZDOCK and 54 . 6% for GRAMM ) as well as the interface residues sampled ( 39 . 6% for ZDOCK and 76 . 8% ) is smaller , which apparently has a strong influence on the method performance ., In particular , the GRAMM docking method performs significantly worse when a subset of docked complexes , consisting only 9-residues is used in the analysis with a % significance of 21 . 6 ±3 . 4 compared to 48 . 4 ±4 . 7 using ZDOCK ., We used 13 different ligand probes and by default 2000 docked conformations to locate the binding site of a receptor protein ., This amounts to 13x2000 = 26 , 000 docked poses ., We gradually reduced the number of docked poses and found that with 13 ligands as few as 200 docked conformations are sufficient to establish the same results as before , with 2000 poses ( Fig 6 ) ., Another aspect of the binding site exploration is the number and variety of probes employed ., Upon plotting the performance of all the 13 probes independently , it is clear that these perform in a relatively tight range and that the observed small differences most likely can be acknowledged to the particular set of test proteins used ., As an empirical test , the accuracy using ZDOCK changes from 65 . 4 when averaged over a subset of 6 randomly selected different probes to 63 . 2 when averaged over all 13 different probes ., We found that randomly selecting 2–3 probes already provides robustly the same performance results as running all 13 probes ( Fig 6 ) ., It has been shown that docking based methods are less successful to predict the correct binding pose and binding site when targeting uncomplexed receptors , especially the ones that undergo substantial conformational change upon binding to their cognate ligand ., In our case we do not restrict our analysis to the cognate ligand and to a few ( or one ) docked poses with the lowest energetics , as such an approach is likely to be insensitive to small conformational changes ., Non-cognate ligands bind with much lower affinity , and we are capturing the relative preference of any ligand to dock to the cognate binding site ., We manually identified 95 target proteins in our combined set for which we could locate a PDB structure in an uncomplexed form ., The F-scores of apo and holo forms for this subset of 95 target proteins is include in the Supplementary Information ( S1 Fig ) ., On this subset , the success rate of capturing binding sites has an average F-score of 0 . 27 and 0 . 26 for the complexed and uncomplexed targets , respectively , a statistically insignificant difference ., An important aspect of this study is to explore if the observed phenomenon is a function of fold types , or something more general ., The distribution of protein folds is very uneven34 , with 12 superfolds populating about one third of the human genome ., It has been discussed in the literature that these superfolds have a tendency to preserve their ancient/general binding interface despite their divergence into a range of distinct functions35 ., We analyzed our dataset to examine whether the well-performing interface detections using unrelated ligand probes work disproportionally well for these superfolds ., Of the 241 protein chains , 91 belong to one of the top 12 CATH47 superfamilies , roughly recapitulating the proportion of superfolds in biological systems ., The success rates for the 91 superfamily and 131 non-superfamily classified cases are 71 . 0% and 62 . 2% using ZDOCK , respectively , and then 30 . 0 and 40 . 8 using GRAMM ( Table 2 ) ., These small and non-systematic differences suggest that there is no preference for superfamily proteins , and that supersites are characteristic to all protein folds ., Further breaking down of the results in a benchmark database dependent fashion shows that the general performance on the Docking Benchmark dataset is significantly better with ZDOCK than with the GRAMM docking approach , and for the NOX dataset these differences are substantially reduced ( Table 3 ) ., However , no systematic preference emerged of supersites in superfolds , in fact , non-superfold subsets outperform in two out of four subsets ( ZDOCK with NOX database , and GRAMM with Docking Benchmark ) ., In order to understand some of the differences in performances , we examined the specific superfamily classifications of the proteins represented in the two datasets ( Table 4 ) ., In the Docking Benchmark , we found a highly skewed distribution of superfolds , where 66% of the superfamily classification is “immunoglobulin-like” while 14% are classified as the Rossman fold ., Meanwhile , the NOX dataset is slightly better balanced , with the Rossman , TIM-barrel , and Immunoglobulin-like folds comprising 48 . 8% , 19 . 5% , and 17% of the dataset , respectively ., It is possible that ZDOCK is better tuned to dock immunoglobulin like folds and their over-representation has shifted the results higher in the Docking Benchmark dataset ., Slightly different interface definitions can drastically change the number of residues involved in the interface ., A recent study suggests that even in the case of nearly identical definitions , the disagreement between different definitions can be substantial , suggesting that a ~0 . 8 F-score as a practical upper limit for prediction methods48 ., In addition , residues not involved in direct contact with a ligand can have a profound effect on binding , as illustrated by a number of studies2 ., Meanwhile , random predictions are distributed with a peak around 0 . 1 F-score28 but many individual random predictions reach up as high as 0 . 2 F-score ., Current protein interface prediction methods that provide results on a residue level and with an F-score accuracy , report statistically significant but generally speaking fairly low accuracies28 , 49 , 50 ., For instance , Table 3 in Taherzadeh et al . 49 published this year , reports seven methods , with F-score performances in the range of 0 . 18–0 . 31 ., These methods typically use different benchmark datasets therefore a substantial part of the variation among the performance probably can be acknowledged to that fact ., To put our results in this general context we converted our performance into F-score evaluation and obtained an average F-score of 0 . 35 using ZDOCK and 0 . 22 using GRAMM , which compares well with the recent values in the literature using other methods to identify protein-protein interfaces ., The good performance is especially promising as our approach is based on the direct evaluation of a single feature while all other methods are using a combination of a number of features in machine learning setting ., In this work , we have shown that protein binding supersites exist in proteins , i . e . the protein binding interface provides an energetically-preferred binding site for many alternative , non-cognate proteins as well ., There were previous , anecdotal studies that noted that even non-cognate ligand have tendency to accumulate around the cognate site , as it was shown in case of chymotrypsin when docked with a non-native binder , lysosyme40 ., Other recent studies also pointed in the direction of our current observation51 , 52 ., Employing an energy landscape based analysis it was observed that binding sites can be identified without the prior knowledge of the cognate ligand ., In that study , in a strict filtering protocol , the few lowest energy binders were identified for subsequent mapping of their preferred binding poses ., Though this approach delivered an effective prediction method , it left open the following question—are these low energy binding poses related to the cognate binding partner , and thereby representing similar binding affinities , and likely , a similar binding interface ?, Also , the observations were not generalized , the successful cases were not analyzed in terms of protein topology , to illustrate if the observations go beyond the original observations made about superfolds , where binding sites are preserved despite a long evolutionary history of sequence divergence ., We observe that these sites can be effectively detected by employing an extensive docking sampling with a range of unrelated protein ligand probes ., In another study the Hex docking approach was used in cross docking experiment and suggested the existence of “favored” sites53 ., The authors have noted a tendency of these sites to be closer to the center of mass of the protein and explored residue type preferences of binding patches ., A wide variety of probes were used with different topologies but the phenomenon was not generalized in terms of distribution on folds , to see if these observations are generic over all fold types or work mostly for superfolds as it was established in 199854 ., The accuracy of this approach to detect protein binding sites is comparable to other state-of-the-art techniques ., However , it uses a mostly orthogonal input in comparison to many existing technologies , and as such , a practical outcome of this study is both a new , standalone binding site prediction algorithm and an opportunity to improve existing binding site predictions by incorporating this information with other existing techniques that use residue preferences , conservation , geometrical definitions , among others ., On the conceptual level , our observations argue that possibly a combination of geometrical restraints ( shape of the local molecular surface ) and energetically preferred residue patterns are responsible for establishing these supersites ., Given past experience and our current results , we believe that the number of combinations of how an energetically “sticky” patch can be established varies substantially ., However , the fact that docking algorithms , which combine shape complementarity with a scoring function that assesses interactions , are able to capture many of these sites suggests a path forward in the characterization of protein interfaces ., Docking methods were benchmarked in a number of studies that showed a lack of strong correlation between calculated and experimental binding affinities55 ., The current study implicitly confirms this observation when we show that the success of identifying binding interfaces does not depend in a statistically significant manner on whether one uses cognate or non-cognate ligands , albeit a small trend favoring cognate ligands can be detected ., This suggests that more generic energetic features are captured ., Two different datasets were employed in this study ., A set of 108 protein chains from the Docking Benchmark44 and another set of 133 protein chains from the NOX database43 , 73 and 60 of which are obligate and non-obligate complexes , respectively ., The protein binding interfaces were identified from the three dimensional structure of the complexes using the CSU56 program ., A residue was considered to be at the interface if any of its atoms is within 3 . 5 Å of any atom of the interacting protein in the complex and establishes a legitimate contact type according to the CSU classification ., In our approach we use a total of 13 ligand probes , none of which are known partners or share any detectable sequence similarity to known ligands for the query proteins in our data set ., Six of these ligand probes were immunoglobulin folds ( PDB57 codes: 1i85 . D , 2jjs . C , 2wbw . C , 1t0p . B , 2ptt . B , 3udw . C ) , as we assumed this fold evolved to be particularly suitable and generic to explore protein surfaces ., Seven others were selected randomly ., PDB entries were split into chains and clustered at 25% sequence identity level ., All protein solved by NMR and not within the range of 70–250 residues were removed ., From the remaining set we selected 7 proteins ( between 70–120 residues ) with different topologies compared to one another ( 1whz . A , 2eaq . A , 2v86 . A , 2w8x . A , 2y2y . A , 3h33 . A , 5cuk . A ) ., Two different docking programs , ZDOCK45 and GRAMM58 were used to generate a maximum of 2000 docked complexes for each of the protein chains in our dataset with each of the 13 ligand probes ., The 2000 complex structures for each receptor-ligand probe pair were analyzed using CSU to identify the residues at the interface , Rik , where i is the residue position number and k is the kth docked complex structure ., If a residue is at the interface , then I ( Rik ) = 1; otherwise , I ( Rik ) = 0 ., A Residue Interface Frequency ( RIF ) , Ni was determined for each residue at position i in the receptor protein by summing over all the 2000 docked structures ., The residues were then ranked based on the Ni values , and the top 15 ranking residues were considered most likely to be at the interface ., The residues were also ranked similarly by using a subset of the 2000 complex structures all of which contained exactly nine residues at the interface ( the most frequent interface patch size during the simulations ) ., This subset generally consisted of between 150 and 300 complex structures ., The actual number of interface residues varies with each receptor protein ., We considered the number of true positive predictions of interface residues in the top 15 rankings assigned by our method ., The performance of the current RIF method was evaluated using a statistical significance test by comparing it with a random prediction ., The probability of randomly selecting x interface residues in the top K predicted residues ( K = 15 in our case ) for a query protein chain with N is the total number of residues sampled during the extensive docking simulation and M actual interface residues is given by the probability mass function of the hypergeometric distribution:, P ( X=x ) = ( Mx ) ( N−MK−x ) / ( NK ), An interface prediction is considered significant if P ( X = x ) < 0 . 05 ., The performance is expressed as, %significance=NumberwithP ( X=x ) <0 . 05TotalnumberinthedatasetX100, Theoretically , the hypergeometric distribution can be exposed to some instability when very small numbers of discrete residues are assessed for significance; therefore , performance was also evaluated empirically , by randomly sampling 15 residues from the surface exposed residues sampled during the extensive docking simulation of the query protein 200 times and finding the average number of interface residues , μ , and the standard deviation , σ ., A Z-score was then calculated , Z = ( N–μ ) / σ , where N is the actual number of interface residues in the top 15 using the RIF method ., The prediction was considered significant if Z > 1 . 97 ., The % significance evaluated using the hypergeometric distribution and the random sampling method yielded identical results ., Receiver operating characteristic curves ROC were calculated by plotting the true positive rate ( sensitivity ) against the false positive rate ( 1- specificity ) ., Corresponding Area Under the Curve values were obtained ., Functional residues are a small fraction of the total residues , so true negatives far outnumber true positives ., Therefore methods that heavily reward true negatives , such as the “specificity” and the “accuracy” , are less appropriate than ones that do not , such as the “F-Score”59 and appropriately F-scores were used in a number of previous studies ., Therefore success of a functional residue prediction was also assessed by the F-score , the harmonic mean of precision and recall ( 2*precision*recall / ( precision + recall ) ) , where precision is the ratio of true positives to the sum of true and false positives and recall is the ratio of true positives to the sum of true positives and false negatives .

Introduction, Results and discussion, Materials and methods

The lack of a deep understanding of how proteins interact remains an important roadblock in advancing efforts to identify binding partners and uncover the corresponding regulatory mechanisms of the functions they mediate ., Understanding protein-protein interactions is also essential for designing specific chemical modifications to develop new reagents and therapeutics ., We explored the hypothesis of whether protein interaction sites serve as generic biding sites for non-cognate protein ligands , just as it has been observed for small-molecule-binding sites in the past ., Using extensive computational docking experiments on a test set of 241 protein complexes , we found that indeed there is a strong preference for non-cognate ligands to bind to the cognate binding site of a receptor ., This observation appears to be robust to variations in docking programs , types of non-cognate protein probes , sizes of binding patches , relative sizes of binding patches and full-length proteins , and the exploration of obligate and non-obligate complexes ., The accuracy of the docking scoring function appears to play a role in defining the correct site ., The frequency of interaction of unrelated probes recognizing the binding interface was utilized in a simple prediction algorithm that showed accuracy competitive with other state of the art methods .

Protein–protein interactions are key to understand the molecular level mechanisms of regulation in the cell ., However , there is still a limited understanding of what distinguishes a protein-protein binding site from the rest of the surface ., This lack of knowledge is manifested in the relatively low accuracy of computational methods that try to predict protein interfaces ., In this work we report a new conceptual insight about protein interfaces ., Our results suggest that protein interfaces serve as generic binding sites to any ligand ., This also means that in the absence of the known binding partner it is still possible to define protein interfaces by extensive docking studies of randomly selected , unrelated ligands , as they have a strong tendency to bind to the cognate binding site ., This insight was leveraged in a new binding interface prediction algorithm that alone outperforms state of the art approaches that often combine a variety of features .

chemical characterization, medicine and health sciences, immune physiology, protein interactions, statistics, immunology, mathematics, forecasting, receptor-ligand binding assay, protein structure prediction, protein structure, antibodies, research and analysis methods, immune system proteins, protein-protein interactions, proteins, mathematical and statistical techniques, binding analysis, molecular biology, biochemistry, biochemical simulations, physiology, biology and life sciences, physical sciences, computational biology, statistical methods, macromolecular structure analysis

journal.pcbi.1005701

Two passive mechanical conditions modulate power generation by the outer hair cells

The mammalian cochlea encodes sounds with pressure levels ranging over six orders of magnitude into neural signals ., This wide dynamic range of the cochlea is achieved by the amplification of low amplitude sounds ., The outer hair cells have been identified as the mechanical actuators that generate the forces needed for cochlear amplification 1 ., Cochlear amplification is dependent on location along the cochlear length ., For example , according to measurements of the chinchilla cochlea , the amplification factor of basilar membrane ( BM ) vibrations was about 40 dB in basal locations while it was 15 dB in apical locations 2–4 ., Theoretical studies have reproduced location-dependent cochlear amplification by adopting tonotopic electrophysiological properties , such as the active feedback gain of the outer hair cells 5 , 6 , or the mechano-transduction properties of the outer hair cell stereocilia 7 , 8 ., These studies are based on experimental reports concerning the tonotopy of the outer hair cells’ electrophysiological properties e . g . , 9 , 10–12 ., On the other hand , recent experimental observations suggest that organ of Corti mechanics may play a role in cochlear amplification ., For example , organ of Corti micro-structures vibrate either in phase or out of phase depending on stimulation level and frequency 13–16 ., These observations challenge a long-standing framework for modeling the organ of Corti mechanics—rigid body kinematics , introduced by ter Kuile 17 ., A fully deformable organ of Corti may have implications for cochlear amplification ., Micro-mechanical aspects of cochlear power amplification were investigated in our previous study , using a computational model of the cochlea 18 ., The model features detailed organ of Corti mechanics analyzed using a 3-D finite element method , and up-to-date outer hair cell physiology ., In that previous work 18 , it was shown that the stiffness of the organ of Corti complex ( OCC ) felt by the outer hair cells remains comparable to the outer hair cell stiffness , independent of location ., An intriguing observation was that even though the same active force gain was used for all outer hair cells , the model reproduced greater amplification toward the base ., However , the specific model aspects responsible for the location-dependence were not identified in that paper ., In this study , by analyzing power generation in individual hair cells , by observing different micro-mechanical transfer functions of the organ of Corti , and through a series of parametric studies , we identify passive mechanical aspects that are responsible for the location-dependent amplification ., Two sets of fluid dynamical , structural and electro-physiological responses are presented in Fig 1 and Fig 2 that represent the active and the passive responses , respectively ., When the stimulating frequency was 18 . 6 , 4 . 4 and 0 . 78 kHz , the BM vibrations of the active cochlea peaked at x = 2 , 6 and 10 mm , respectively ( Fig 1 ) ., For the same stimulating frequencies , the peak responding location shifted toward the base when passive ( x = 1 . 2 , 5 . 4 , 9 . 7 mm , Fig 2 ) ., This shift of peak responding location due to the outer hair cell active feedback is consistent with experimental observations e . g . , 19 , and other model studies e . g . , 20 , 21 ., The pressure amplitude plots show one or two peaks and notches as the pressure propagates from the oval window toward the apex ., This is similar to the pressure peaks/notches observed experimentally by Kale and Olson 22 ., As was discussed in their work , the pressure peaks/notches were generated by the interference between fast and slow pressure components ., For example , the slow ( differential pressure across the OCC ) component showed no pressure notch ., When the amplification is large ( Fig 1A ) , the local pressure near the peak responding location was large enough to mask the pressure pattern created by the oval window motion ., Despite changes in the outer hair cell active force , the pressure at the stapes remains relatively constant at about 80 mPa for a 1 nm/ms stapes velocity amplitude ., After considering the stapes footplate area of 0 . 8 mm2 23 , our simulated result corresponds to a cochlear input impedance of 100 GPa·s/m3 ., This value is comparable to measured values ranging between 50 and 300 GPa·s/m3 24–26 ., Explicit computation of the interactions between the outer hair cells and OCC fine structures is both an opportunity and a challenge of our continuum mechanics-based approach ., Any poorly determined parameters can affect the result , and there are a large set of model parameters ., However , as more OCC mechanical data accumulate , well defined parameters can serve as rigorous constraints on the model ., For example , the geometrical information of the OCC is well known , but underused ., The elastic moduli or stiffness of different OCC structures have been measured as summarized in 27 ., This study takes advantage of such existing data ., As a result , the vibration amplitude ratios and phase relationships between the micro mechanical structures vary depending on location , simulating frequency , and active force feedback ., The resulting micro-mechanical responses compare well with experimental observations ., For example , the vibration pattern ( the relative motion between different structural components ) changes depending on the outer hair cell’s active feedback: when active , the TM ( tectorial membrane ) vibrations lead the BM vibrations by 15 to 60 degrees ( Fig 1B ) , but they vibrate in phase when passive ( Fig 2B ) ., This dependence of vibration patterns on the outer hair cell motility has been observed experimentally in the gerbil cochlea 13 ., Our model predicts that the micro-mechanical response characteristics are location-dependent: at middle to apical locations , the TM vibrates less than the BM , but the opposite is true in the basal turn ( x < 4 mm ) ., The relative motion between the BM and other OCC structures , caused by the outer hair cell’s active feedback , indicates that the top and the bottom of the OCC are effectively decoupled ., In the field of vibration measurement , the vibration patterns due to internal forces are referred to as the operational deflection shapes ., This decoupling due to the outer hair cell action persists along the entire cochlear length 28 ., Up-to-date physiological properties of outer hair cell mechano-transduction and electromotility are incorporated into our model to predict the electro-mechanical feedback of the cells to acoustic stimulations ., The amplitudes of the mechano-transduction current and receptor potential are presented in Fig 1D , and Fig 2D ., Independent of location or outer hair cell feedback force , mechano-transduction current is nearly in phase with the BM displacement ., Using these first-hand results ( fluid pressure , vibration amplitudes of micro-structures , and electrical responses of the outer hair cells ) , we analyzed how the outer hair cells’ power generation is modulated by the organ of Corti mechanics ., Model responses at three different locations ( x = 9 , 6 , and 3 mm ) are presented together with experimental results in Fig 3 ., There are measurements of the gerbil cochlear vibrations at different locations e . g . , 19 , 29 ., Cooper and Rhode’s experiment with the chinchilla cochlea 4 is also pertinent to this study , because the mechanical amplification measurements from different cochlear locations were reported in a single paper ., Our simulated results for two key quantitative measures that represent cochlear performance are in reasonable agreement with experimental observations 4 , 19 , 29 ., First , the BM vibrations are amplified by 30–50 dB in high-frequency locations ( x < 4 mm ) , and 10–20 dB in low-frequency locations ( x > 8 mm ) ., Second , the tuning quality represented by Q10dB are > 3 in the base , and ~1 in the apex ., While gain and phase curve shapes are in reasonable agreement with the experiment , there are differences between our model response and experimental results in absolute values ., As compared to Ren and Nuttall’s measurements 19 , the gain was lower by 17 dB , and the phase was different by a half cycle ., The difference in gain may be ascribed to different reference input conditions: our kinematic boundary condition at x = 0 and 0 < y < H , does not exactly represent the stapes excitation ., The half a cycle phase difference between simulation and experimental results could be due to different definitions of positive stapes displacement ., In this study , the stapes velocity is positive when moving into the cochlea ., The phase of BM vibration versus stimulating frequency has been extensively measured and analyzed because it characterizes the cochlear traveling waves e . g . , 30 , 31 ., Three key characteristics are reproduced by our simulations in Fig 3B: First , there exist more than two cycles of total phase accumulation as the frequency increases ., Second , the phase accumulation at the peak responding location is between 1 and 3 cycles ., Finally , the slope of the phase versus frequency curve is similar for the active and passive cases ., The amplification and tuning quality of the cochlea decrease toward the apex ( Fig 3C and 3D ) ., Because the amplification is the primary consequence of the outer hair cell’s active feedback , the location-dependent amplification implies that the outer hair cells provide more power in the base ., One possible approach to model this location-dependent amplification is to assume that the outer hair cells in the basal cochlea have greater active force gain than those in the apex ., Alternatively , the passive mechanics of the OCC may be responsible for the location-dependent amplification ., Note that our model adopted a constant force gain ( gOHC of 0 . 1 nN , active force per mV membrane potential change after 11 ) independent of location ., We investigated which location-dependent properties could be responsible for the greater amplification in the base ., To investigate the origin of the location-dependent amplification , power generation by individual outer hair cells was analyzed ( Fig 4 ) ., The power provided by an outer hair cell to its external system is defined as the product of the active force generated by the cell ( fOHC ) and the rate of cell’s length change ( vOHC , see Eq ( 6 ) in Methods ) ., In Fig 4A , fOHC and vOHC at a moment of time are presented for high , mid and low frequency stimulations ( 18 . 6 , 4 . 4 , and 0 . 78 kHz ) ., For a stapes velocity of 1 nm/ms , the peak values of fOHC and vOHC are 0 . 65 nN and 1 . 2 mm/s at 18 . 6 kHz , and 0 . 11 nN and 21 μm/s at 0 . 78 kHz ., This trend of decreasing force and velocity toward lower stimulating frequency is consistent with decreased amplification toward the apex ., The timing ( phase ) of outer hair cell force generation also contributes to the location-dependent amplification in the cochlea ., The phase of vOHC with respect to fOHC depends on stimulating frequency ( Fig 4B ) ., At the peak-responding location , fOHC lags vOHC by 17 , 61 and 83 degrees for stimulating frequencies of 18 . 6 , 4 . 4 and 0 . 78 kHz , respectively ., For given fOHC and vOHC amplitudes , the power generation by an outer hair cell is greatest when fOHC is in phase with vOHC , and zero when they are 90 degrees out of phase ., In other words , the fOHC-vOHC phase causes the outer hair cells in the base to be 8 times more efficient in generating power than those in the apex ( cos ( 17° ) /cos ( 83° ) ≈ 8 ) ., The power generation per cycle of individual outer hair cells ( computed from Eq ( 6 ) in Methods ) is shown in Fig 4C ., For a stapes vibration amplitude of 1 nm/ms , an individual outer hair cell at the peak responding location generates 377 , 12 , and 0 . 36 fW for the three stimulating frequencies of 18 . 6 , 4 . 4 and 0 . 78 kHz , respectively ., The apical value is comparable to values reported by Wang et al . 32 , but the basal value is about two to three orders of magnitude greater ., This discrepancy could be due to the difference in model species ( Wang et al . modeled the mouse cochlea ) ., As Wang et al . discussed , small difference in vibration amplitude can result in different estimations of outer hair cell power generation ( i . e . , the gerbil cochlea may vibrate greater than the mouse cochlea ) ., Alternatively , the difference may be ascribed to difference in model assumptions such as dissipating mechanisms ., While we lumped the effect of power loss due to the viscous fluid with the viscous damping of the OCC structures , Wang and her colleagues incorporated the fluid viscosity explicitly and considered the damping within the OCC negligible ., Further investigation of the OCC mechanical impedance may be required to better understand the difference ., The power flux ( computed using Eq ( 1 ) in Methods ) represents how much energy is transferred by the scala fluid along the cochlear length ., In the passive system , the power is provided through the stapes ., As a result , the longitudinally transmitted power is dissipated as the traveling waves propagate ., That is , the power flux decreases monotonically toward the apex when passive ( dashed curves , Fig 4D ) ., However , for the active cochlea ( comparable to the case of weak sound stimulation to healthy ears ) , the power flux pattern is non-monotonic—the power flux increases until it culminates at the best responding location ( solid curves , Fig 4D ) ., A similar trend of power flux was shown in other theoretical studies 32 , 33 ., In Figs 1 and 2 , it was shown that the TM and BM vibrate in phase when passive , but the TM leads the BM displacement by 15 to 60 degrees when active ., A plausible theory is that the active feedback of the outer hair cells modulates the OCC mechanics to facilitate outer hair cell power generation ., We examined whether the outer hair cell active feedback also modulates the phase between fOHC and vOHC ., Different levels of amplification were simulated using different values for the active gain gOHC ( between 0 and 0 . 1 nN/mV , constant through the cochlear length ) ., The amplification level and the phase between fOHC and vOHC were analyzed ( Fig 5 ) ., As expected , the power generated by the outer hair cell , and the level of amplification increases as gOHC increases ( Fig 5A ) ., The phase between fOHC and vOHC is minimally affected by the level of outer hair cell active feedback ( Fig 5B ) ., For high values of gOHC , the basal location is amplified more than the apical location ., This location-dependent amplification is consistent with the phase relationships between fOHC and vOHC: fOHC is approximately in phase with vOHC in the basal location , but fOHC is roughly in phase with the outer hair cell displacement in the apex ( panel B ) , regardless of amplification level ., According to this result , the active feedback does not ‘correct’ the phase relationship toward more favorable amplification ., That is , the fOHC-vOHC phase becomes less favorable for amplification as the active gain increases ., We investigated which aspect of OCC micro-mechanics is responsible for the location-dependent phase relationship ., The amplitudes and phases of six variables with respect to the BM displacement are shown in Fig 6 ., They are the transverse, ( y ) and radial, ( z ) displacement of the TM , the mechano-transduction current and receptor potential of the outer hair cells , and the stereocilia and somatic displacement of the outer hair cells ., These responses were obtained at various distances from the base , for each location’s best-responding frequency ., The following observations were made: First , the gain of most variables decreases toward the apex ( top panels of Fig 6 ) ., Second , the phases of most variables are approximately flat over the distance ( bottom panels of Fig 6 ) ., Third , these gain and phase trends are similar with and without outer hair cell feedback ( i . e . , the plots of the left and the right columns are similar ) ., Many of the mechanical responses were approximately in phase with the BM vibrations ( within ±10 degrees ) when the mechanics were passive ( gray curves in Fig 6B bottom panel ) , and were up to 60 degrees out of phase when active ( gray curves in Fig 6A bottom panel ) ., There are two variables that deviate from the general trends ., The gain of the TM radial displacement ( zTM , curves with □ markers ) varies non-monotonically with distance from the base ., The gain of zTM has a minimum value near x = 4 mm , and its phase relative to the BM shifts 180 degrees near this location ., Although the 180 degree-shift is reminiscent of a resonator , it is not due to resonance ., For example , a change of TM stiffness or mass does not change the characteristic location of the phase shift ., An explanation for this phase shift is given in the next section ., The phase of the outer hair cell somatic displacement ( dOHC , curves with ° markers ) increases toward the apex by about 90 degrees with or without outer hair cell feedback ., Considering that the phase of the receptor potential ( Vm , curves with * markers ) remains near -60 degrees over the distance when active , the dOHC phase is primarily responsible for the location-dependent phase difference between fOHC and vOHC in Fig 5 , since vOHC = jωdOHC ., Note that the trend of the dOHC phase with distance is not affected by the active feedback of the outer hair cells ., To conclude , the extent of the outer hair cell power generation is passive mechanically regulated ., The vibration pattern of the OCC represented by the phase of dOHC is more favorable for amplification in basal locations ., In the present model , the point of attachment between the TM and the spiral limbus is above the reticular lamina in the base ( indicated by the dimension e , Fig 7A ) , but it is below the reticular lamina in the apex ( Fig 7B ) ., This variation of the TM attachment geometry at different locations is modeled after available anatomical data of the gerbil cochlea 34 , 35 ., For example , the length of an inner pillar cell is greater than the height of the TM attachment ( c > e ) in the apex , but the opposite is true in the base ( c < e ) 35 ., The variation of the zTM phase with respect to the BM displacement is determined by geometry , specifically by the attachment angle of the TM ( θ in Fig 7 ) ., The TM motion trajectories are shown in Fig 7—the red and black curves for active and passive simulations , respectively ., The blue curve normal to the BM is the BM trajectory to which the TM trajectory is referenced ., For example , the TM to BM amplitude ratio is greater in the basal location than in the apical location ., In the basal location , the TM to BM amplitude ratio is greater when active than when passive ., The opposite is true in the apical location ., The BM vibrates minimally in the radial direction ., Roughly speaking , the TM rotates about its attachment point ., Because the TM is subject to axial deformation in addition to bending ( rotational ) deformation , the TM motion trajectory is not exactly normal to the TM ., When the OCC kinematics is dominated by the bending deformation , the sign of the attachment angle ( θ ) determines the direction of zTM ., The Meaud-Grosh model is consistent with this interpretation in that it incorporates the bending and the axial motions of the TM explicitly 20 , 36 ., Further investigation is necessary to learn the functional implications of the bending and axial motion of the TM ., In Fig 7A and 7B , the TM radial displacements are in the positive and negative directions , respectively , for the peak BM upward displacement ., In agreement with experimental observations , an angle difference as small as 10 degrees can result in this approximately 180-degree phase reversal of zTM ., Although this TM geometry affects the nominal direction of the TM radial motion ( zTM ) , we do not find a functional consequence of the geometry , i . e . , the zTM phase is not correlated with parameters related to cochlear amplification such as the phase of dOHC ., Our model is complex with many independent parameters ., Different parameter sets can result in similar functional characteristics , including the level of cochlear amplification ., To identify parameters critical for cochlear amplification , a series of sensitivity analyses were performed ( Fig 8 ) ., With only one model parameter altered from its standard value , amplification factors were obtained ., Active and passive responses to a 4 kHz pure tone were used to define the amplification factor ., For 4 kHz stimulation , the traveling waves peak in the middle of the model ( ~6 mm from the base ) ., The elastic properties of the outer hair cell’s body and hair bundle affect the amplification more prominently than other supporting structures ( Fig 8A ) ., As expected , as the level of damping increases the amplification decreases ( Fig 8B ) ., Note that this study uses inviscid fluids ., Any viscous dissipation in the fluids or in the OCC is approximated with the Rayleigh damping term ., The active gain of the somatic motility ( gOHC ) has a strong effect on the amplification , but the active gain of stereocilia motility ( gMET ) has negligible effect on the amplification ( Fig 8C ) ., This suggests that , in the present model , the somatic motility is the primary active component for amplification ., This series of sensitivity analyses reveals that there exist different sets of parameters that result in the same amplification level ., For example , the same level of amplification can be achieved by reducing both OCC damping and active gain , or by decreasing outer hair cell stiffness while increasing OCC damping ., The outer hair cell stiffness , the damping imposed on the BM , and the active gain of the outer hair cells are the three most sensitive parameters in our model ., As the model parameter increases by a factor of two near the standard value , the amplification level changes approximately by -20 dB , -10 dB and +30 dB , for the outer hair cell stiffness , OCC damping and outer hair cell active force gain , respectively ., Consistent with a previous study ( Meaud , Grosh , 2011 ) and our previous report ( Liu et al . , 2015 ) , the active hair bundle force minimally affects the OCC mechanics ., The somatic force is one to two orders of magnitude greater than the hair bundle force according to available physiological data ( Nam , Fettiplace , 2012 ) ., For a 1 nm BM vibration amplitude , the fOHC amplitude ranged between 172 and 11 pN and the fMET amplitude ranged between 13 and 0 . 17 pN , over the range of x between 2 and 10 mm ., The minimal contribution of fMET to amplification may be ascribed to its small force magnitude ., However , even when the active force gain of fMET ( fMET , max in Table 1 ) was increased by a factor of 10 , the contribution of hair bundle forces to amplification remained minimal ., This suggests that the somatic force is more favorably situated to deliver power for cochlear amplification ., Although we did not observe a significant effect of fMET on amplification , conditions may exist when the hair bundle force can modulate the OCC mechanics more effectively ( e . g . , Ó Maoiléidigh , Hudspeth , 2013 ) ., In theory , the trend of greater amplification in higher frequency locations can be reversed by adjusting model parameters ., To determine which parameters have the greatest effect on the location-dependent amplification , we attempted to reverse the location-dependent amplification trend ., Although damping or active gain can also affect the trend , we could not find a set of values that completely reverses the amplification trend by adjusting only these parameters ., In contrast , the location-dependent amplification trend is readily modulated by adjusting the relative stiffness of the OCC felt by individual outer hair cells ., The stiffness of the OCC felt by each outer hair cell has consequences for cochlear amplification 18 , 36 ., It is the relative stiffness of the OCC as compared to the outer hair cell stiffness that is relevant to power generation by the outer hair cells 37 , 38 ., To compute the relative OCC stiffness , a set of equal-and-opposite forces was applied to the outer hair cell body ( fC ) or the stereocilia ( fB ) , and the corresponding static displacements ( δC or δB ) were obtained ( Fig 9A and 9B ) ., The relative OCC stiffness felt by an outer hair cell is defined as rOHC = ( fC/δC—kOHC ) /kOHC , where kOHC and fC/δC—kOHC are the stiffness of the outer hair cell body and the stiffness of the OCC felt by the outer hair cell , respectively ., Likewise , the relative OCC stiffness felt by the outer hair cell bundle is defined as rOHB = ( fB/δB—kOHB ) /kOHB , where kOHB and fB/δB—kOHB are the stiffness of the outer hair cell bundle and the stiffness of the OCC felt by the outer hair cell hair bundle , respectively ., We introduce the parameters rOHC and rOHB for two reasons ., First , the stiffness of outer hair cell or its stereocilia bundle has a much greater effect on cochlear amplification than other OCC structures ( Fig 8 ) ., Second , according to the theory of impedance matching 37 , 38 , it is the impedance ratio between an actuator and the actuated system that determines the efficiency of power transmission ., We hypothesized that the trend of greater amplification toward the base is a consequence of rOHC or rOHB monotonically decreasing toward the apex ., To test this hypothesis , while all other model parameters remained the same , either kOHC or kOHB was adjusted so that the longitudinal trend of rOHC or rOHB was reflected with respect to the center at x = 6 mm ( curves with dot symbols , Fig 9C and 9F ) ., kOHB or kOHC was adjusted instead of other OCC mechanical properties , because a change in the other OCC structures such as the BM or TM stiffness can change the fundamental tonotopy ( i . e . , the definition of base and apex will become obscured if the BM stiffness is reversed ) ., Although we simulated a wide range of outer hair cell stiffness values , it is not because their properties are poorly grounded ., The mechanical properties of the outer hair cell and the stereocilia are better understood than other fine structures in the organ of Corti ., For example , the hair bundle stiffness has been measured to be ~3 mN/m for 4–5 μm tall outer hair cell hair bundles 39 ., Our standard properties ( 40 and 4 . 5 mN/m for 2 and 6 μm-tall hair bundles , respectively ) are within a reasonable range ., The axial stiffness of the outer hair cell body has been measured to be 500 nN per unit strain independent of location 11 ., Our standard property is 950 nN per unit strain ., We used conservative ( greater ) kOHC and kOHB values than the measured values after considering experimental factors that could influence measured values e . g . , 40 ., To reverse the longitudinal trend of rOHC or rOHB , the ratio between kOHB values at x = 2 and 10 mm is increased from 9 to 140 or the kOHC ratio is increased from 2 . 4 to 50 ., These large variations in stiffness seem unlikely , but are used here as a simple way to illustrate the effect of rOHC and rOHB ., Reversing the longitudinal profile of rOHC or rOHB results in a reversed amplification trend along the cochlear length ( Fig 10 ) ., As a result of this adjustment , the active tuning curves at three locations show less amplification and blunt tuning in the base , greater amplification and sharp tuning in the apex ( Fig 10 A and 10D ) ., In contrast to active responses , passive responses were affected minimally by the reversal of rOHC or rOHB ( Fig 10B and 10E ) ., The amplification level versus location shows that the location-dependent amplification trend is reversed with the adjusted rOHC or rOHB ( Fig 10C and 10F ) ., The OCC transfer functions of the test cases with a reversed longitudinal pattern of rOHC or rOHB reveal that the two parameters modulate cochlear amplification differently ., Figs 11 and 12 present how reversed rOHC and rOHB affect the OCC transfer functions ., Reversing the spatial pattern of rOHC affected the outer hair cell length change ( the curves with filled circles , top panels of Fig 11 ) , but barely affected the phase relationship ., Reversing the spatial pattern of rOHB affected both the amplitude and phase of the radial TM motion with respect to the BM motion ( the curves with ■ , Fig 12 ) ., The characteristic phase shift of zTM near x = 4 mm disappears because the TM no longer behaves like a rigid bar hinged at the attachment point when the hair bundle stiffness becomes comparable to or greater than the TM axial stiffness ( Fig 7C ) ., Despite similar outcomes , the mechanisms by which rOHC and rOHB affect cochlear amplification are different ., Fig 13 summarizes the differences ., The change of rOHC affects the amplification in two ways ., First , as the stiffness of the outer hair cell body increases , the effective active force that is used to deform the OCC other than the cell itself decreases ., Second , because the outer hair cells act as an elastic coupler between the TM and the BM , the modulation of their stiffness affects the vibration pattern ., When the longitudinal dependence of rOHC is reversed , the phase between vOHC and fOHC becomes more favorable for apical power generation as compared to the standard case ( the curve with ○ , Fig 13A ) ., Unlike the case of reversed rOHC , the reversed rOHB hardly affected the vOHC-fOHC phase relationship ( the curve with ■ , Fig 13A ) ., The change of rOHB affected the amplification by changing the mechanical gain of the hair bundle displacement ( Fig 13B ) ., For example , as a result of decreased kHB ( or increased rOHB ) , the hair bundle displacement gain ( dHB/yBM ) was increased from 0 . 25 to 0 . 55 at x = 10 mm ., Because dHB is a part of the loop determining the active feedback gain , doubling the hair bundle gain ( dHB/yBM ) is comparable to doubling the active gain ( gOHC ) ., According to Fig 8C , doubling gOHC resulted in an approximately 40 dB increase in amplification ., Unlike the case of reversed rOHB , reversing rOHC hardly affects the hair bundle displacement gain ., Because location-dependent amplification is a well-known characteristic of cochlear physiology , theoretical studies have reproduced this characteristic for validation ., However , the origin of location-dependent amplification is quite different from theory to theory , revealing the lack of agreement for the mechanism of cochlear amplification ., There are three major aspects to consider:, 1 ) the active force gain of the outer hair cell;, 2 ) the limiting speed of outer hair cell force generation , often referred to as the RC time constant issue; and, 3 ) the timing ( phase ) of active force application ., Some theoretical studies assumed that the force gain of cochlear actuators varies with location ., For example , Mammano and Nobili 5 made two assumptions—the outer hair cell force cancels the damping of BM vibrations , and its amplitude is proportional to the BM stiffness ( their Eq ( 10 ) ) ., Lu et al . 6 used a gain that exponentially varies over the cochlear length to represent the outer hair cell force ( parameter kf in their Table 1 ) ., The location-dependent amplification of these studies may represent a graded capacity of the feedback force with phase-locked force application ., Although our study used a constant outer hair cell electro-mechanical gain ( gOHC of 0 . 1 nN/mV ) , when referenced to the BM displacement like the previous studies , the mechanical and electrical gain of the present model varies with location ( Fig 6 ) ., In that sense , our study is not inconsistent with the previous studies ., Instead , our study divides the active gain into two components: the OCC mechanical gain , and the electro-mechanical gain of the outer hair cells ., In this study , we demonstrated that the OCC mec

Introduction, Results, Discussion, Methods

In the mammalian cochlea , small vibrations of the sensory epithelium are amplified due to active electro-mechanical feedback of the outer hair cells ., The level of amplification is greater in the base than in the apex of the cochlea ., Theoretical studies have used longitudinally varying active feedback properties to reproduce the location-dependent amplification ., The active feedback force has been considered to be proportional to the basilar membrane displacement or velocity ., An underlying assumption was that organ of Corti mechanics are governed by rigid body kinematics ., However , recent progress in vibration measurement techniques reveals that organ of Corti mechanics are too complicated to be fully represented with rigid body kinematics ., In this study , two components of the active feedback are considered explicitly—organ of Corti mechanics , and outer hair cell electro-mechanics ., Physiological properties for the outer hair cells were incorporated , such as the active force gain , mechano-transduction properties , and membrane RC time constant ., Instead of a kinematical model , a fully deformable 3D finite element model was used ., We show that the organ of Corti mechanics dictate the longitudinal trend of cochlear amplification ., Specifically , our results suggest that two mechanical conditions are responsible for location-dependent cochlear amplification ., First , the phase of the outer hair cell’s somatic force with respect to its elongation rate varies along the cochlear length ., Second , the local stiffness of the organ of Corti complex felt by individual outer hair cells varies along the cochlear length ., We describe how these two mechanical conditions result in greater amplification toward the base of the cochlea .

The mammalian cochlea encodes sound pressure levels over six orders of magnitude ., This wide dynamic range is achieved by amplifying weak sounds ., The outer hair cells , one of two types of receptor cells in the cochlea , are known as the cellular actuators that provide power for the amplification ., It is well known that high frequency sounds encoded in the basal turn of the cochlea are amplified more than low frequency sounds encoded in the apical turn of the cochlea ., This difference in amplification has been ascribed to a difference in electrophysiological properties , such as the membrane capacitance and conductance of the outer hair cells at different locations ., Whether the outer hair cells have a sufficient range of electrophysiological properties to explain the location dependent amplification has long been a topic of scientific debate ., In this study , we present an alternative explanation for how the low and high frequency sounds are amplified differently ., Using a detailed computational model of the cochlear epithelium ( the organ of Corti ) , we demonstrate that the micro-mechanics of the organ of Corti can explain the variation of amplification with longitudinal location in the cochlea .

velocity, stiffness, mechanical properties, medicine and health sciences, classical mechanics, vibration, ears, membrane potential, electrophysiology, neuroscience, outer hair cells, organ of corti, inner ear, materials science, cellular structures and organelles, animal cells, head, cell membranes, physics, cellular neuroscience, cell biology, anatomy, cochlea, physiology, neurons, biology and life sciences, cellular types, afferent neurons, physical sciences, material properties, motion

journal.ppat.1006389

Mycobacterium tuberculosis arrests host cycle at the G1/S transition to establish long term infection

The mechanisms whereby Mycobacterium tuberculosis ( Mtb ) senses the host environment to maintain metabolic homeostasis to establish infection are poorly understood ., Metabolic homeostasis of any cell is sustained by bioenergetic pathways , such as respiration and glycolysis , which provide the cell’s energy requirements in the form of ATP ., In the lung , which is the site of infection in pulmonary tuberculosis ( TB ) , it was found that when the lung macrophages were depleted as a result of acute infection , the majority of repopulation occurred by stochastic cellular proliferation in situ in a macrophage colony-stimulating factor ( CSF ) and granulocyte macrophage-CSF dependent manner 1 ., Interleukin-4 has also been shown to induce an increase in resident macrophage numbers beyond homeostatic levels without coincident monocyte recruitment nor increased recruitment of inflammatory cells 2 ., Further studies 3 , 4 suggest that macrophage proliferation contributes to normal tissue homeostasis and that macrophages can replicate at the site of inflammation ., There is also evidence for in vivo alveolar macrophage proliferation 5 , 6 ., Thus , the proliferation of tissue resident lung macrophages in TB will be predisposed to modulation by Mtb ., Mtb alters essential host functions by the release of polyketides , lipids and cell wall components such as poly- and di-acyltrehaloses ( PAT/DAT ) , phosphatidylinositol mannosides 1 & 2 ( PIM 1 , 2 ) and 6 ( PIM6 ) , trehalose dimycolate ( TDM ) , sulfolipids ( SL-1 ) , phenolic glycolipids ( PGL-1 ) , mycolic acids and phthiocerol dimycocerosates ( PDIM ) during infection 7 ., Mtb PhoP 8 and WhiB3 9 are key regulators of these lipids ., Other bacteria secrete or directly inject effector molecules and toxins into the host that interfere with the eukaryotic cell cycle to facilitate disease or persistence; these effector molecules have been termed cyclomodulins 10 ., Cyclomodulins can be inhibitory , for example , the cytolethal distending toxin ( CDT ) produced by Escherichia coli , Shigella dysenteriae and Salmonella typhi blocks the host cell cycle at the G2/M transition 11 and the vacuolating cytotoxin ( VacA ) of Helicobacter pylori induces G1 cell cycle arrest and cell death 12 ., Stimulatory cyclomodulins , including E . coli cytotoxic necrotizing factors , Bordetella dermonecrotic toxin , and H . pylori CagA promote cell proliferation 11 ., Cyclomodulins are not always proteins as is evident by the production of the polyketide mycolactone by Mycobacterium ulcerans 13 and E . coli polyketides 14 ., Polyketides are lipid-like molecules that are smaller than known protein toxins but have potent biological activities , for example , antibiotic ( erythromycin ) , immunosuppressant ( rapamycin ) and antifungal ( amphotericin B ) ., Previously , we have shown how Mtb WhiB3 affects the virulence of two pathogenic mycobacterial strains in different animal models 15 ., WhiB3 is a 4Fe-4S cluster DNA binding protein that maintains intracellular redox balance by sensing host-generated NO and O2 16 and modulating virulence polyketide lipids to cause disease 9 ., As the cytoplasmic redox environment is tightly coupled to central metabolism , WhiB3 was implicated in regulating the mycobacterium’s metabolism ., Here we hypothesized that WhiB3 maintains bioenergetic homeostasis to control production of factors that subvert host cell function ., To test this hypothesis , we exploited a combination of Mtb and macrophage transcriptomic analyses , real-time bioenergetic flux analysis , and a series of host cell cycle analyses ., To investigate mechanisms of WhiB3-mediated virulence , we examined the global transcriptome of wt Mtb and MtbΔwhiB3 grown in 7H9 media to mid-log phase ( Fig 1 ) ., For a complete list of the 315 WhiB3 regulated genes , see S1 Table ., Notably , WhiB3 controls the expression of 50 genes involved in intermediary metabolism and respiration , of which 11 genes are concerned directly with aerobic respiration and energy metabolism ( e . g . , ctaE , atpB , atpF and atpH , lldD2 , gltA ) ., The down-regulation of ATP synthase subunits and a component of the aa33-type cytochrome oxidase c ( CtaE ) in MtbΔwhiB3 , all essential components of oxidative phosphorylation , suggest that WhiB3 is involved in regulating bioenergetic homeostasis ., Downregulation of CtaE , which functions as an H+ pump , would reduce the extrusion of protons across the membrane during respiration strongly suggesting a role for WhiB3 in maintaining an energized membrane , which is essential for bioenergetic homeostasis ., Conversely , the gene encoding citrate synthase 3 ( gltA-1 ) of the methylcitrate cycle , which generates succinate that feeds into succinate dehydrogenase , SDH , of the electron transport chain ( ETC ) , is upregulated in MtbΔwhiB3 ., This suggests that the mutant is attempting to restore an energized membrane potential by upregulating the formation of succinate ., As previously demonstrated by our studies 9 , the transcriptomic data ( Fig, 1 ) indicates that WhiB3 is involved in redox homeostasis with the downregulation of genes encoding: antioxidants , such as thioredoxin ( trxC ) ; electron carriers , such as ferredoxin ( fdxC ) ; enzymes , such as nitrate reductase ( narH ) , other reductases ( hemA , cysH , nrdE ) , monooxygenases ( rv1393c ) and enzymes that utilize reducing equivalents in anabolism ( sulfite reductases , sirA , and glutamine synthase , glnA1 ) ., Not surprisingly , the down-regulation of a large number ribosomal genes ( Fig 1 ) , which is essential for protein synthesis and the major consumer of cellular energy 17 , 18 , is consistent with a role of WhiB3 in maintaining bioenergetic homeostasis ., A noticeable observation was that WhiB3 negatively regulates 43 of 56 insertion sequence ( IS ) loci , representing almost 30 ISs ( Fig 1 ) ., Earlier studies have shown that intermediary metabolism , redox balance , and DNA metabolism controls transposition of ISs 19 ., These studies have shown that the formation of transposomes is facilitated by nucleoid-associated proteins such as HNS , HU and IHF 19 ., Intriguingly , WhiB3 regulates several nucleoid associated proteins including HNS , Lsr2 and HupB ( Fig 1 ) , which suggests a role for WhiB3 in the regulation of IS ., A previous transcriptomic study investigating the role of WhiB3 in acid resistance of Mtb 20 also found WhiB3 regulated genes involved in redox homeostasis , secretion and lipid metabolism differentially regulated in response to acidic pH; although there were only 12 significantly regulated genes in common with our transcriptomic data , primarily involved in secretion and lipid metabolism ., This is expected as our microarray was performed at neutral pH , in contrast to Mehta et al . investigating differential regulation between acidic pH ( 4 . 5 ) and neutral pH . WhiB3 also regulates the expression of several genes belonging to the ESX-1 secretion system including , but not limited to esxA , esxB , espA , espC and espD , pointing to a complex role in the regulation of virulence ., In line with the WhiB3 regulation of virulence lipid anabolism previously published by Singh et al . ( 2009 ) , genes encoding polyketide synthases ( pks16 , 3 , 4 and 2 ) , other enzymes involved with polyketide synthesis ( papA3 , papA1 ) and lipid anabolism ( mmaA4 , fbpB , cdh , fbpA ) were downregulated in MtbΔwhiB3 ., The microarray data were validated by performing Q-PCR on select genes regulated by WhiB3 ( S2 Table ) ., In sum , the genes involved in intermediary metabolism and respiration , lipid metabolism , translation , transposition and secretion provide fresh insight into the role WhiB3 plays in virulence ., In support of WhiB3 regulation of bioenergetics ( Fig 1 ) , the ATP concentration in MtbΔwhiB3 was found to be significantly lower ( p<0 . 0001 ) than that in wt Mtb and the complemented strain over a five-day growth period ( Fig 2B ) , despite all three strains having matching growth rates ( Fig 2A ) ., NADH feeds electrons into the electron transport chain at complex I and is essential for bioenergetic homeostasis ., Using LC-MS/MS , significantly higher NADH/NAD+ ratios were observed in MtbΔwhiB3 in comparison to wt and the complemented strain when the bacilli were grown in the presence of physiological concentrations of cholesterol , palmitate , glucose , or no carbon sources for 24 h whereas no significant differences were noted when grown on acetate or propionate ( Fig 2C ) ., The increased NADH/NAD+ ratio in MtbΔwhiB3 grown in glucose correlates with the lower ATP levels in the WhiB3 mutant ., This reaffirmed that WhiB3 maintains the intrabacterial NADH/NAD+ poise , but demonstrates it in light of the ability of WhiB3 to regulate the metabolic switch in response to available carbon sources 9 ., Also , it indicates that WhiB3 indirectly regulates the donation of electrons into the ETC by controlling intracellular levels of NADH that donate electrons to menaquinone via Complex I ( NADH dehydrogenase ) into the ETC to sustain oxidative phosphorylation ( OXPHOS ) ., The influence of WhiB3 on mycobacterial bioenergetics was determined by examining extracellular flux of Mtb , MtbΔwhiB3 and the complemented strain using a XF96 Extracellular Flux Analyzer 21 ., This analyzer allows continuous real-time quantification of O2 consumption and extracellular acidification of multiple samples with high sensitivity in a non-invasive manner 21 ., The mycobacteria were adhered to each well in specialized microplates and fluorophore-based O2 and H+ sensors measured their oxygen consumption rate ( OCR ) in pmol O2 per minute , which gives a measure of OXPHOS , and extracellular acidification rate ( ECAR ) in mpH units per minute , which represents carbon catabolism , respectively ( Fig 2D ) ., Plots of OCR versus ECAR , known as phenograms , give an overall illustration of the energy phenotype of the cells ., Initially , extracellular flux of the three strains was measured in 7H9 media without a carbon source ( CS ) and after the addition of glucose , acetate , or propionate ., No significant changes were observed in the OCR or ECAR before and after the addition of the carbon source to all three strains , apart from glucose , where there was a slight increase in OCR of MtbΔwhiB3 after the addition of glucose ( Fig 2E–2H ) ., To address the capacity of the mycobacteria to cope under bioenergetic stress when growing in different CSs , increasing concentrations of the membrane uncoupler , CCCP , were added ., When no CS was present , the whiB3 mutant had no capacity to increase OCR in response to the bioenergetic stress induced by CCCP ( Fig 2E ) ., In the presence of glucose or acetate , after the addition of 8 μM CCCP , OCR and ECAR of the MtbΔwhiB3 increased significantly more than OCR and ECAR of the wt and complemented strains ( Fig 2F and 2G ) ., This is particularly noticeable in the phenograms of glucose and acetate ., However , in the presence of propionate ( Fig 2H ) , no differences were observed among all three strains after the addition of 8 μM CCCP ., This provides evidence for a role of WhiB3 in regulating the metabolic switch and in turn energy metabolism in response to available environmental carbon sources in the presence of bioenergetics stress ., In sum , the microarray , ATP concentrations , NADH/NAD+ ratios and the extracellular flux analysis suggest Mtb WhiB3 is necessary for maintaining bioenergetic equilibrium ., Using methods developed for microarray analysis of intracellular Mtb 22 , transcript levels of intracellular Mtb and MtbΔwhiB3 24 h post-infection of bone-marrow derived macrophages ( BMDM ) were compared to extracellular controls to reveal relative changes in gene expression in response to the phagosomal environment ( Fig 3A and S3 Table ) ., Secondly , we directly compared the transcript levels of intracellular Mtb and MtbΔwhiB3 24 h post-infection of BMDM ( Fig 4 and S4 Table ) ., Distinct gene expression profiles were noted in bacilli from resting versus activated macrophages , consistent with previous reports 23 ., However , analysis of the data indicated that the macrophage activation status did not significantly affect differential regulation of genes to the extent of that observed between Mtb and MtbΔwhiB3 under our experimental conditions ( Fig 3B ) , thus we excluded the effect of macrophage activation status ., Furthermore , Rohde et al . 22 found that WhiB3 was induced by a decrease in phagosomal pH , which occurs when macrophages are activated by IFN-γ or lipopolysaccharide , or when resting macrophages are infected with Mtb ., In the first approach , we observed differential regulation of 333 genes following invasion of BMDM that was significantly dependent on WhiB3 ( S3 Table ) ., We found a striking influence of WhiB3 loss on genes involved in intermediary metabolism and respiration ( 62 genes ) in addition to genes implicated in the cell wall and cell processes , such as cell wall transporters ( 60 genes ) ( Fig 3A ) pointing to the roles of WhiB3 in the adaptation of Mtb to the phagosome ., Genes involved in glycolysis ( pgk ) , the citric acid cycle ( gltA2 ) , methyl citrate cycle ( rv1130 , prpD ) , gluconeogenesis ( pckA ) , and respiration ( ndh ) were all upregulated in MtbΔwhiB3 ., Genes involved in amino acid metabolism , in particular for His and Leu were upregulated in MtbΔwhiB3 , while genes involved in the sulphur metabolism ( including cysteine metabolism , cysa2 , metK , rv1464 ) were downregulated ., While mutA was upregulated in the methyl malonyl pathway , genes encoding cobS and cobU in the biosynthesis of Vitamin B12 ( a cofactor for mutA/B ) were downregulated in MtbΔwhiB3 ., This data suggests WhiB3 , which is induced by phagosomal acidification 22 , regulates metabolism upon infection to enable Mtb to adapt to the intraphagosomal environment ., Evidence for the role of WhiB3 in maintaining an energized membrane potential necessary for bioenergetic homeostasis is the whiB3-dependent expression of a large number of cell wall ion transporter genes: the magnesium and cobalt transport protein ( corA ) and molybdate permease ( modB ) , which are both upregulated in MtbΔwhiB3; transporting ATPases , including cation ( ctpG which is upregulated and ctpD which is downregulated in MtbΔwhiB3 ) , anion ( rv3680 , which is downregulated in MtbΔwhiB3 ) , potassium ( kdpB ) and magnesium ( mgtC ) , which are both upregulated in MtbΔwhiB3 ., Other transporters modulated by WhiB3 included ABC transporters ( rv1348 ) , and transport system proteins for peptides ( dppA , dppC ) , dicarboxylate ( ktpG ) , sugar ( uspC ) and phosphate ( phoY1 ) ., The differential regulation of these cell wall transporters also indicates the role of WhiB3 in response to phagosome acidification and maintaining intrabacterial pH . WhiB3 also controls the transcription of 25 genes involved in lipid metabolism ( Fig 3A ) in response to phagosomal cues , notably , those involved in polyketide synthesis , pks2 and pks3 were downregulated while pks5 was upregulated , in addition to genes implicated in lipid degradation ( fadA3 and fadD5 were downregulated , fadD8 and fadE5 were upregulated ) , mycolic acid metabolism ( acpM , fbpB , inhA , mmaA4 were all upregulated ) and mycobactin synthesis ( mbtB , mbtC , mbt1 were all upregulated . ) ., WhiB3-regulated expression of genes involved in lipid metabolism suggests WhiB3 is involved in altering the cell wall composition in response to the hostile phagosomal environment as well as preparing for the establishment of a persistent infection aided by these lipid virulence factors ., Additionally , a large number of genes involved in redox homeostasis , including oxidoreductases , dehydrogenases , monooxygenases and genes involved in sulphur metabolism were differentially transcribed confirming the role of WhiB3 in maintaining intracellular redox balance ., Previous transcriptomic studies investigating the role of WhiB3 in acidic pH ( 4 . 5 ) resistance of Mtb to elucidate mechanisms Mtb uses to survive in the phagosomes during immune activation 20 revealed similar overlaps to our intraphagosomal transcriptomic data in the classes of genes regulated by WhiB3 , namely , redox homeostasis , amino acid metabolism and lipid metabolism ., However , among the genes significantly differentially regulated in these classes , there were only 6 genes in common between the two transcriptomic studies ., Whereas genes involved in secretion featured in the acidic transcriptomic data , genes involved in cell wall transporters were regulated by WhiB3 in our intraphagosomal transcriptomic data ., In the second analysis , 121 genes were upregulated and 101 downregulated in intracellular MtbΔwhiB3 on comparison to intracellular Mtb ( S4 Table ) ., Genes involved in intermediary metabolism , respiration , the cell wall and cell processes were the most markedly influenced by WhiB3 ( Fig 4 ) demonstrating that WhiB3 is involved in regulating these pathways to protect Mtb and potentially alter the hostile environment in the phagosome ., Genes involved in the TCA Cycle ( gltA2 , sucC ) and amino acids feeding into the TCA cycle ( glnE , argC , argJ ) , Vitamin B12 biosynthesis , a cofactor for the methylmalonyl pathway ( cobU ) , respiration ( ctaD , menB , hemZ ) , glycolysis ( pgmA ) and its link to glycerol metabolism ( glpD2 ) were downregulated in MtbΔwhiB3 in the phagosome ., In contrast , genes involved in nucleotide and deoxyribonucleotide catabolism ( deoC ) were upregulated in MtbΔwhiB3 to produce glyceraldehyde-3-phosphate for glycolysis ., Changes in the central carbon metabolism of the mutant mycobacterium suggest that WhiB3 is pivotal in altering metabolism to ensure intracellular survival ., Genes encoding cell wall transporters ( ctpA and fecB ) and enzymes involved in cell wall remodeling ( cut3 , cut4 , rv1984c ) were downregulated in MtbΔwhiB3 indicating that WhiB3 is involved in the continual adjusting of the cell wall to the changing intraphagosomal environment ., Twenty-four genes involved in lipid metabolism were also regulated by WhiB3 in the intraphagosomal environment ., Of importance , genes involved in polyketide synthesis ( pks2 , pks3 , pks4 , papA3 ) and sulfolipid-1 ( papA1 ) and methoxymycolic acid synthesis ( mmA2 ) were downregulated in MtbΔwhiB3 ., Taken together , the intracellular expression data demonstrate that WhiB3 plays a role in the continuous adaptation of Mtb to the phagosome environment by adjusting respiration , central carbon metabolism and transporters in the cell wall together with cell wall remodeling ., Importantly , as previously observed 9 , WhiB3 also plays a role in virulence by regulating the synthesis of immunomodulatory polyketides in macrophages ., To gain insights into how the activity of WhiB3 might impact host cell function , we compared the global transcription profiles of replicating RAW264 . 7 macrophages infected with Mtb or MtbΔwhiB3 ., Using Affymetrix Mouse 430 2 . 0 arrays , 45 037 probe sets were examined of which 629 host genes were differentially regulated by Mtb versus MtbΔwhiB3 ( S5 and S6 Tables ) ., To elucidate the biological significance of these genes , we made use of MetaCoreTM and identified the top 10 significant host pathways that are differentially regulated in MtbΔwhiB3 infected macrophages , of which seven are involved in the host cell cycle: including: DNA damage check points , biomechanical stress , cytoskeletal remodeling , chromosome condensation and apoptosis ( Fig 5A ) ., RAW264 . 7 macrophages are capable of proliferation with a doubling time of 11 hours and hence a complete cell cycle ., Subsets of differentially regulated critical genes involved in four pathways are listed in S1 Fig . S2 Fig generated by MetaCoreTM , illustrates the most significant pathways under WhiB3 control involved in the cell cycle regulation of G1/S transition and cytoskeletal rearrangement ., These gene network findings were supported by subsequent LC-MS/MS proteomic analyses of uninfected , Mtb and MtbΔwhiB3 infected RAW264 . 7 macrophages , which also revealed differential regulation of proteins involved with the cell cycle between uninfected and Mtb or MtbΔwhiB3 infected macrophages ( Fig 5B and S7 Table ) ., In Mtb infected macrophages , proteins such as: nuclear ubiquitous casein and Cdk substrate 1 ( NUCKS1 ) 24 and nuclear factor related to kappa-B-binding protein ( NFRKB ) 25 , which are involved in genomic stability , homologous recombination and the DNA repair pathway; the protein S100-A1 , a calcium binding protein involved in inhibition of microtubule assembly 26; mitogen activated protein kinase 9 ( MAPK9 , also called JNK2 ) involved in regulating the exit from G1 of the cell cycle 27 , 28; Cdc42-interacting protein 4 ( Trip10 ) that promotes cell death or survival in a cell-dependent manner 29 , were upregulated when compared to MtbΔwhiB3 infected macrophages ., These proteins are all involved in retaining the cells in the G0-G1 phase of the cell cycle ., Conversely , proteins upregulated in MtbΔwhiB3 infected macrophages when compared to Mtb infected macrophages , include cyclin dependent kinase 4 ( CDK4 ) that initiates the cell cycle through the G1-S transition 30 , CDK7 , which activates CDK4 31; protein S100A4 that drives cells into the G2/M phase; proteins that attenuate apoptosis and promote proliferation , such as CDK15 32 and protein S100A10 33 , 34; and BRACA2 and CDKN1A-interaction protein ( BCCIP ) , which has paradoxical roles in that it has been shown to delay G1-S progression but is also a perquisite for proliferation 35 , 36 ., NUCKS is probably downregulated in the Mtb infected cells after 48 hours as DNA repair proteins are not only essential for maintenance of genomic integrity , but often fundamental in DNA replication and mitosis 35 that is possibly being inhibited by the Mtb infection ., The roles of the upregulation of the cyclin-dependent kinase-like 1 and the downregulation of growth arrest-specific protein 7 ( Gas7 ) in Mtb infected macrophages are unclear at this stage ., In summary , both transcriptomic and proteomic analysis of infected RAW264 . 7 macrophages implicate WhiB3 in regulating the host cell cycle ., Based on transcription profiles in S5 and S6 Tables , we hypothesize that factors controlled by Mtb WhiB3 will affect host cell DNA synthesis , proliferation and differentiation as the host cell cycle controls the stages at which these events occur ., WhiB3 modulation of macrophage DNA synthesis in the cycling mouse macrophage cell line , RAW264 . 7 , was confirmed by fluorescence microscopy of macrophages infected with Mtb , MtbΔwhiB3 and complemented strains for 24 h in the presence of 5-bromo-2-deoxyuridine ( BrdU ) , a thymidine analogue that is incorporated into DNA during replication and detected with an anti-BrdU antibody ., The data revealed greater BrdU incorporation and thus increased DNA synthesis in MtbΔwhiB3 infected RAW264 . 7 macrophages than in macrophages infected with wt and complemented strains ( Fig 5C ) ., As macrophage proliferation contributes to normal tissue homeostasis 3 and is present at sites of inflammation 4 , we investigated the effect of MtbΔwhiB3 infection on immune cell proliferation in vivo ., We used BrdU incorporation assays to assess DNA synthesis , leading to host cell proliferation in Mtb , MtbΔwhiB3 and complemented infected mice ( Fig 5D–5F ) ., After the mouse lungs were harvested , the single cell suspension was labelled with anti-CD3 ( lymphocyte marker ) , anti-CD11b and anti-CD11c ( both myeloid cell markers ) and anti-BrdU ., Flow cytometry revealed a significantly higher number of CD3+ , CD11b+ or CD11c+ cells were positive for BrdU incorporation in the mice infected with MtbΔwhiB3 than the mice infected with wt or the complemented strain , demonstrating increased DNA synthesis in host immune cells ( Fig 5E ) ., Furthermore , the immune cell markers revealed a significantly increased number of total CD11c+ cells in the lungs of mice infected with MtbΔwhiB3 , indicating proliferation in the mouse lungs infected with MtbΔwhiB3 ( Fig 5F ) ., This increased CD11c+ cell proliferation in the MtbΔwhiB3 infected lung was significantly greater than the proliferation observed in the CD11c+ cells in the lungs of mice infected with wt ., This suggests that Mtb WhiB3 is involved in modulating the natural proliferative immune response in mice that was observed when infected with MtbΔwhiB3 ., In conclusion , these findings propose that Mtb WhiB3 plays a role in immunomodulation by suppressing DNA synthesis and CD11c+ cell proliferation ., Eukaryotic cell division proceeds through a regulated cell cycle comprising of five phases: G0 , the resting phase , G1 ( the normal growth phase ) , S ( DNA replication phase ) , G2 ( growth and preparing for mitosis ) , and the M ( mitosis ) phase ( Fig 6A ) , of which the DNA content can be accurately determined using propidium iodide ( PI ) , and flow cytometry 37 ., BrdU incorporation in addition to PI staining emphasizes new DNA synthesis and gives a more improved representation of the S phase ( Fig 6B ) ., RAW264 . 7 macrophages have been used in previous studies to examine cell cycle modulation induced by pathogen infection 38 , 39 , and we used these macrophages to generate the transcriptomic data ( Fig 5A ) ., Thus , for correlation purposes , cell cycle progression in uninfected and Mtb , MtbΔwhiB3 or complemented strain infected macrophages was monitored using PI staining and BrdU incorporation at 12 , 24 , 36 and 48 h post-infection ( Fig 6C ) ., Notably , in comparison to uninfected macrophages , there was a significant increase in Mtb infected macrophages in the G0/G1 phase at 24 and 36 h ( Fig 6D ) , followed by a significant reduction of cells in the S phase ( Fig 6E ) as well as in the G2 phase ( Fig 6F ) at all time points monitored ., Similar trends were observed in the macrophages infected with the complemented strain ., This suggests that Mtb is prolonging the G1 phase of the infected macrophages and reducing their entry into the S-phase ., This confirms the host transcriptomic data where the cell cycle regulation of G1/S transition is the most differentially regulated pathway ( Fig 5A ) ., In addition , the cell cycle of macrophages infected with MtbΔwhiB3 resembles that of the uninfected cells at all time points investigated ( Fig 6D , 6E and 6F ) , suggesting that the host cell cycle was not significantly affected by the MtbΔwhiB3 infection ., In sum , these findings suggest that Mtb inhibits the G1/S transition in the macrophage cell cycle and WhiB3 is involved in this modulation ., Here , we hypothesize that mycobacterial polyketides and cell surface lipids , some under WhiB3 control , regulate the host cell cycle ., Hence , we examined the effect of purified extracts of polyketides and cell surface lipids on the RAW264 . 7 macrophage cell cycle ( Untreated , Fig 7M ) ., The effects of the solvents used to dissolve the lipids on the macrophage cell cycle were assessed and regarded as controls ( Fig 7A , 7D and 7G ) ., Of the lipids examined ( Fig 7B , 7C , 7E , 7F , 7H and 7I ) , mycolic acid methyl esters ( MAME ) had the most potent effect on the host cell cycle ., MAME , PDIM-1 , PIM 1 , 2 , total lipids and TDM ( Fig 7B , 7C , 7E , 7H and 7I ) significantly enhanced the percentage of macrophages in the G1 phase ( p≤0 . 005 ) together with significant reductions of that in the S and G2 phases ., Higher concentrations of PDIM ( Fig 7C ) did not have significant effects ( p>0 . 05 ) on the host cell cycle , probably due to aggregation of PDIM ., PIM 1 , 2 ( Fig 7E ) also significantly increased the percentage of apoptotic cells , indicated by the sub-G1 population ( p<0 . 01 ) ., Lower concentrations of PIM 1 , 2 ( Fig 7E ) significantly lowered cell numbers in the G2 phase with no significant effects on the G1 and S-phases ., SL-1 ( Fig 7F ) modulated the host cell cycle with significant increments of cells in the G1 and S phases , and a significant decrease of cells in the G2 phase ( p<0 . 05 ) ., Lower concentrations of SL-1 ( Fig 7F ) significantly increased the number of apoptotic cells ( p<0 . 005 ) ., When the effects of the polyketides and lipids at identical concentrations , 1 μg/ml , on RAW264 . 7 macrophages over 24 h were compared , all the lipids examined induced a significant increase in the number of cells in the Sub-G1 ( apoptotic cells ) and G0-G1 phase , with a significant concomitant decrease in percentage of cells in the S phase and the G2 phase when compared to the untreated cells and vehicle ( 0 . 1% DMSO ) -treated cells ( Fig 7J ) ., In a physiological setting , these lipids would be present in combinations , hence we examined the cell cycles of macrophages treated with a combination of SL-1 and PDIM , and a combination PIM1 , 2 and PDIM for 24 h in opposing and equal concentrations ( Fig 7K , 7L , 7N and 7O ) ., Regardless of the varied concentrations of the lipids examined , both combinations resulted in significant increases in the percentage of cells in the G0-G1 phase and significant reductions in the number of cells in the S and G2-M phases suggesting that the combination of lipids also prohibited the macrophage cell cycle transition from the G1 to the S phase ., In sum , we found that mycobacterial lipids do modulate the RAW264 . 7 macrophage cell cycle by inhibiting the G1-S transition ., To further validate our findings that mycobacterial lipids under WhiB3 control modulated the host cell cycle , we investigated if the infection of RAW264 . 7 macrophages with Mtb mutants for polyketides under WhiB3 control demonstrated similar alterations of the host cell cycle as MtbΔwhiB3 ., Macrophages were infected with Mtb , Mtb Tn:pks2 ( necessary for SL-1 production ) , Mtb Tn:mas and Mtb Tn:ppsA ( both required for PDIM production ) for 12 , 24 , 36 and 48 h ., Using BrdU and PI incorporation ( Fig 8A ) , we found that macrophages infected with Mtb Tn:pks2 , Mtb Tn:mas or Mtb Tn:ppsA had significantly greater numbers of cells in the S-phase ( Fig 8C ) , with concomitant increments of cells in the G2-M phase ( Fig 8D ) and fewer cells in the G0-G1 phase ( Fig 8B ) at all time points investigated in comparison to that in wt infected macrophages ., These cell cycle modulations are similar to the trends in the host cell cycle observed in uninfected and MtbΔwhiB3 infected macrophages ( Fig 6D , 6E and 6F ) ., Thus , the parallels observed in the host cell cycle progression between macrophages infected with the whiB3 mutant and those infected with the polyketide mutants strongly suggests that the polyketides under WhiB3 control are responsible for altering the host cell cycle ., A major challenge in the TB field is to understand the precise mechanism of disease to develop novel therapeutic strategies against Mtb pathogenesis ., Mtb WhiB3 has emerged as a model virulence redox regulator , although the exact mechanism is unknown ., Initially , we investigated mechanisms of WhiB3 virulence using transcriptomic analysis of Mtb and infected macrophages ., In vitro and intraphagosomal Mtb microarrays indicated the WhiB3 dependence of bioenergetic metabolism and its role in enabling Mtb to adapt to the phagosomal environment by cell wall remodeling and metabolic modulation , in particular lipid metabolism , amino acid metabolism , redox and sulfur metabolism ( Figs 3 and 4 ) ., Unexpectedly , the host microarray data revealed that Mtb WhiB3 differentially regulated host genes involved in the host cell cycle and DNA damage checkpoints in macrophages that are capable of cell division ( Fig 5 ) ., Although there is the supposition that macrophages are terminally differentiated , there is clear evidence for in vivo lung macrophage proliferation in the lung 1 , which is the site of infection in pulmonary TB ., Thus , the proliferating lung macrophages will be predisposed to modulation by Mtb lipids and cell wall components after ingestion of Mtb ., We subsequently hypothesized that Mtb WhiB3 maintains bioenergetic homeostasis to control production of factors that subvert host c

Introduction, Results, Discussion, Methods

Signals modulating the production of Mycobacterium tuberculosis ( Mtb ) virulence factors essential for establishing long-term persistent infection are unknown ., The WhiB3 redox regulator is known to regulate the production of Mtb virulence factors , however the mechanisms of this modulation are unknown ., To advance our understanding of the mechanisms involved in WhiB3 regulation , we performed Mtb in vitro , intraphagosomal and infected host expression analyses ., Our Mtb expression analyses in conjunction with extracellular flux analyses demonstrated that WhiB3 maintains bioenergetic homeostasis in response to available carbon sources found in vivo to establish Mtb infection ., Our infected host expression analysis indicated that WhiB3 is involved in regulation of the host cell cycle ., Detailed cell-cycle analysis revealed that Mtb infection inhibited the macrophage G1/S transition , and polyketides under WhiB3 control arrested the macrophages in the G0-G1 phase ., Notably , infection with the Mtb whiB3 mutant or polyketide mutants had little effect on the macrophage cell cycle and emulated the uninfected cells ., This suggests that polyketides regulated by Mtb WhiB3 are responsible for the cell cycle arrest observed in macrophages infected with the wild type Mtb ., Thus , our findings demonstrate that Mtb WhiB3 maintains bioenergetic homeostasis to produce polyketide and lipid cyclomodulins that target the host cell cycle ., This is a new mechanism whereby Mtb modulates the immune system by altering the host cell cycle to promote long-term persistence ., This new knowledge could serve as the foundation for new host-directed therapeutic discovery efforts that target the host cell cycle .

Mycobacterium tuberculosis ( Mtb ) is responsible for the estimated 1 . 8 million people with tuberculosis ., One of the reasons for the success of this pathogen is its ability to modulate the immune system and establish a persistent infection ., The manner whereby the mycobacterium senses the environment and modulates the host immune system is poorly understood ., In this study , we used transcriptional analyses of both Mtb and the infected macrophage to ascertain mechanisms whereby Mtb adapts to and resides in macrophages ., We found that WhiB3 , a redox sensor in Mtb that controls virulence lipid production , is also involved in modulating the mycobacterium’s energy metabolic pathways in response to available carbon sources ., As redox homeostasis regulates the virulent lipid production in Mtb , and the oxido-reductive homeostasis is tightly coupled with bioenergetic homeostasis , the virulent lipid production will be dependent on bioenergetic homeostasis ., From the host’s perspective , transcriptional analysis revealed that Mtb regulates the macrophage’s cell cycle and comprehensive cell cycle analysis indicated that Mtb arrested the macrophages’ cell cycle ., We discovered that polyketides under Mtb WhiB3 control were responsible for this cell cycle arrest that will potentially modulate the immune response to this intracellular pathogen ., These studies reveal a novel strategy of targeting the host cell cycle for chemotherapeutic intervention .

blood cells, medicine and health sciences, immune cells, respiratory infections, viral transmission and infection, cell cycle and cell division, immunology, cell processes, microbiology, pulmonology, physiological processes, homeostasis, bioenergetics, bacteria, lipids, white blood cells, animal cells, actinobacteria, biochemistry, host cells, cell biology, mycobacterium tuberculosis, virology, physiology, biology and life sciences, cellular types, macrophages, organisms

journal.pcbi.1003246

dPeak: High Resolution Identification of Transcription Factor Binding Sites from PET and SET ChIP-Seq Data

Since its introduction , chromatin immunoprecipitation followed by high throughput sequencing ( ChIP-Seq ) has revolutionized the study of gene regulation ., ChIP-Seq is currently the state-of-the-art method for studying protein-DNA interactions genome-wide and is widely used 1–5 ., ChIP-Seq experiments capture millions of DNA fragments ( in length ) that the protein under study interacts with using random fragmentation of DNA and a protein-specific antibody ., Then , high throughput sequencing of a small region ( ) at the end or both ends of each fragment generates millions of reads or tags ., Sequencing one end and both ends are referred to as single-end tag ( SET ) and paired-end tag ( PET ) technologies , respectively ( Figure 1A ) ., Standard preprocessing of these data involves mapping reads to a reference genome and retaining the uniquely mapping ones 6 , 7 ., In PET data , start and end positions of each DNA fragment can be obtained by connecting positions of paired reads 8 ., In contrast , the location of only the end of each DNA fragment is known in SET data ., The usual practice for SET data is to either extend each read to its direction by the average library size which is a parameter set in the experimental procedure 7 or shift the end position of each read by an estimate of the library size 9 ., Then , genomic regions with large numbers of clustered aligned reads are identified as binding sites using one or more of the many available statistical approaches 6 , 7 , 9–11 ( the first step in Figure 1C ) ., Currently , the SET assay dominates all the ChIP-Seq experiments despite the fact that PET has several obvious , albeit less studied , advantages over SET ., In PET data , paired reads from both ends of each DNA fragment can reduce the alignment ambiguity , increase precision in assigning the fragment locations , and improve mapping rates ., This is especially advantageous for studying regulatory roles of repetitive regions of genomes 12 , 13 ., Although many eukaryotic genomes are rich in repetitive elements , PET technology has not been extensively used with eukaryotic genomes 8 , 14 ., One of the main reasons for this is that ChIP-Seq data is information rich even when the repetitive regions are not profiled 15 and that the PET assay costs times more than the SET assay ., Put differently , given a fixed cost , PET sequencing results in a lower sequencing depth compared to SET sequencing ., In contrast to eukaryotic genomes , prokaryotic genomes are highly mappable , e . g . , of the Escherichia coli ( E . coli ) genome is mappable with reads ., This decreases the higher mapping rate appeal of the PET assay for these genomes ., In this paper , we systematically investigate advantages of the PET assay from a new perspective and demonstrate both experimentally and computationally that it significantly improves the resolution of protein binding site identification ., Improving resolution in identifying protein-DNA interaction sites is a critical issue in the study of prokaryotic genomes because prokaryotic transcription factors have closely spaced binding sites , some of which are only to apart from each other 16–19 ., These closely spaced binding sites are considered to be multiple “switches” that differentially regulate gene expression under diverse growth conditions 17 ., Therefore , identification and differentiation of closely spaced binding sites are invaluable for elucidating the transcriptional networks of prokaryotic genomes ., Although many methods have been proposed to identify peaks from ChIP-Seq data ( reviewed in 20 ) , such as MACS 9 , CisGenome 6 , and MOSAiCS 10 , these approaches reveal protein binding sites only in low resolution , i . e . , at an interval of hundreds to thousands of base pairs ., Furthermore , they report only one “mode” or “predicted binding location” per peak ., More recently , deconvolution algorithms such as CSDeconv 21 , GPS 22 ( recently improved as GEM 23 ) , and PICS 11 have been proposed to identify binding sites in higher resolution ., However , these methods are specific to SET ChIP-Seq data and are not equipped to utilize the main features of PET ChIP-Seq data ., Although a relatively recent method named SIPeS 24 is specifically designed for PET data and is shown to perform better than MACS paired-end mode 9 , our extensive computational and experimental analysis indicated that this approach is not suited for identifying closely located binding events ., To address these limitations , we developed dPeak , a high resolution binding site identification ( deconvolution ) algorithm that can utilize both PET and SET ChIP-Seq data ., The dPeak algorithm implements a probabilistic model that accurately describes the ChIP-Seq data generation process and analytically quantifies the differences in resolution between the PET and SET ChIP-Seq assays ., We demonstrate that dPeak outperforms or performs competitively with the available SET-specific methods such as PICS , GPS , and GEM ., More importantly , dPeak coupled with PET ChIP-Seq data improves the resolution of binding site identification significantly compared to SET-based analysis with any of the available methods ., Generation and analysis of factor PET and SET ChIP-Seq data from E . coli grown under aerobic and anaerobic conditions reveal the power of the dPeak algorithm in identifying closely located binding sites ., Our study demonstrates the importance of high resolution binding site identification when studying the same factor under diverse biological conditions ., We further support our findings by validating a small subset of our closely located binding site predictions with primer extension experiments ., The factor is responsible for transcription initiation at over 80% of the known promoters in E . coli 25 ., combines with RNA polymerase to bind promoter sequences typically containing two consensus elements located at and upstream of the transcription start site 18; thus a binding site spans about upstream from the transcription start site ., Many E . coli genes contain multiple promoters , and much transcriptional regulation by oxygen as well as by other stimuli occurs by selection of one or a subset of the possible promoters in concert with binding of activators and repressors ( e . g . , ArcA and FNR for regulation by oxygen 17 , 19 ) ., Understanding such regulation requires knowledge of precisely which promoters are used in a given condition ., Therefore , the highest possible accuracy of ChIP-signal mapping will allow the best determination of promoter binding by -RNA polymerase holoenzyme ., We generated both PET and SET ChIP-Seq data for factor from E . coli grown under aerobic ( ) and anaerobic ( ) conditions in glucose minimal media on the HiSeq2000 and Illumina GA IIx platforms ., We used these experimental data for comparisons of PET and SET assays and evaluation of our high resolution binding site detection method dPeak throughout the paper ., Figure 1B displays PET and SET ChIP-Seq coverage plots for the promoter region of the cydA gene under the aerobic condition ., The height at each position indicates the number of DNA fragments overlapping that position ., The cydA promoter contains five known binding sites separated by to 25 ., As evidenced in Figure 1B , coverage plots for PET and SET appear almost indistinguishable visually ., To further understand the appearance of peaks that multiple binding events in this region would result in , we simulated PET and SET data with parameters matching to those of this region ., Figures S1A , B , C in Text S1 display SET and PET coverage plots of this region when it harbors one and three binding events ., These plots support that when binding events are in close proximity with distances less than the average library size , they appear as uni-modal peaks regardless of the library preparation protocol ( Figure S1C in Text S1 ) ., We next evaluated two peak callers , MACS 9 and MOSAiCS 10 , both of which are specifically developed for SET data , on our SET and PET experimental datasets ( Table S1 in Text S1 ) ., Both methods identified broad regions and the median widths of MACS peaks were to times larger than those of the MOSAiCS peaks ., Detailed comparison of the MACS and MOSAiCS peaks revealed that each MACS peak on average has to MOSAiCS peaks ( Table S2 in Text S1 ) ., Next , we evaluated the number of annotated binding events from RegulonDB 25 ( http://regulondb . ccg . unam . mx/ ) in each of the MACS and MOSAiCS peaks and found that MACS peaks , on average , had to annotated binding events whereas MOSAiCS peaks had to ., Overall , we did not observe any differences in the peak widths of the PET and SET assays with MOSAiCS whereas MACS peaks from PET data tended to be wider than those of the SET data ., These findings indicate that the potential advantages of the PET assay for elucidating closely located binding sites are not simply revealed from visual inspection and by analysis with methods developed specifically for SET data ., Hence , deciphering the advantages of PET over SET for high resolution binding site identification warrants a statistical assessment ., Next , we developed a generative probabilistic model and an accompanying algorithm , dPeak , that can specifically utilize local read distributions from SET and PET assays ., This algorithm enabled unbiased evaluation of the SET and PET assays using our E . coli SET and PET ChIP-Seq data ., dPeak requires data in the form of genomic coordinates of paired reads ( for PET ) or genomic coordinates of reads and their strands ( for SET ) obtained from mapping to a reference genome ., For computational efficiency , dPeak first identifies candidate regions ( i . e . , peaks ) that contain at least one binding event and considers each candidate region separately for the prediction of number and locations of binding events ( the first step of Figure 1C ) ., Either two-sample ( using both ChIP and control input samples ) or one-sample ( only using ChIP sample when a control sample is lacking ) analysis can be used to identify candidate regions ., For this purpose , we utilize the MOSAiCS algorithm 10 which produced narrower peaks than the MACS algorithm 9 in our ChIP-Seq datasets ( Table S1 in Text S1 ) ., In each candidate region , we model read positions as originating from a mixture of multiple binding events and a background component ( the third step of Figure 1C ) ., dPeak infers the number of binding events and the read sets corresponding to each binding event within each region ., It iterates the following two steps for each candidate region ., First , it assigns each read to a binding event or background , based on the positions and strengths of the binding events ., Then , the position and strength of each binding event are updated using its assigned reads ., In practice , the number of binding events in each candidate region is unknown a priori ., Hence , we consider models with different numbers of binding events and choose the optimal number using Bayesian information criterion ( BIC ) 26 ., We constructed generative probabilistic models for binding event components and a background component for each of the PET and SET data by careful exploratory analyses of multiple experimental ChIP-Seq datasets ., Diagnostic plots of the fitted models ( Figure S3 in Text S1 ) indicate that the dPeak model fits ChIP-Seq data well ., dPeak has two unique features compared to other peak deconvolution algorithms ( Table S3 in Text S1 ) ., First , it accommodates both SET and PET data and explicitly utilizes specific features of both types ., Second , it incorporates a background component that accommodates reads due to non-specific binding ., Consideration of non-specific binding is critical because the degree of non-specific binding becomes more significant as the sequencing depths get larger ., An additional unique feature of dPeak is the treatment of unknown library size for SET data ., As discussed earlier , to account for unknown library size , each read is either extended to or shifted by an estimate of the library size in most peak calling algorithms 20 ., This estimate is often specified by users 7 , 10 or estimated from ChIP-Seq data 9 , 11 ., Currently available algorithms with the exception of PICS use only one extension/shift estimate for all the regions in the genome ., However , our exploratory analysis of real ChIP-Seq data and the empirical distribution of the library size from PET data ( Figure S2A in Text S1 ) indicate that using single extension/shift length might be suboptimal for peak calling ( data not shown ) ., In order to address this issue , dPeak estimates optimal extension/shift length for each candidate region ., Comparison of empirical distribution of the library size from PET data with the estimates of the region-specific extension/shift lengths indicates that dPeak estimation procedure handles the heterogeneity of the peak-specific library sizes well ( Figures S2B , C , D in Text S1 ) ., This advancement ensures that dPeak is well tuned for deconvolving SET peaks , which then enables an unbiased computational comparison between the SET and PET assays ., We compared dPeak with two competing algorithms , GPS 22 and PICS 11 , for analysis of SET ChIP-Seq data ., We did not include the CSDeconv algorithm 21 in this comparison because it is computationally several orders of magnitude slower than the algorithms considered here ., We utilized the synthetic ChIP-Seq data which was previously used to evaluate deconvolution algorithms 22 ., In this synthetic data , binding events were generated by spiking in reads from predicted CTCF binding events at predefined intervals 22 without explicitly implanting binding sequence motifs ., Therefore , we also excluded GEM 23 , which capitalizes on motif discovery to infer positions of binding events , from this comparison and used additional computational experiments below to perform comparisons with GEM ., The synthetic data from 22 consisted of 1 , 000 joint ( i . e . , close proximity ) binding events , each with two events , and 20 , 000 single binding events ., We assessed performances of algorithms on these two sets separately ., Figure 2A shows the sensitivity of each algorithm at different distances between the joint binding events ., Here , sensitivity is the proportion of regions for which both of the two true binding events are correctly identified ., dPeak outperforms other methods across all considered distances between the joint binding events and especially for closely located binding events separated by less than the average library size of ., When the distance between the joint binding events is about , dPeak is able to identify both binding events in of the regions whereas neither PICS nor GPS can detect both binding events in more than ., Further investigation indicates that PICS merges closely spaced binding events into one event too often ( Figure S4 in Text S1 ) ., We also found that GPS estimates the peak shape incorrectly when ChIP-Seq data harbors many closely located binding events ( Figure S5 in Text S1 ) ., Furthermore , the sensitivity of GPS also decreases significantly when the distance between joint binding events increases ., A closer look at the results reveals that GPS filters out too many predictions for joint binding events ., To ensure that increased sensitivity of dPeak is not a result of increased number of false predictions , we evaluated positive predictive value ( fraction of predictions that are correct ) of each method ., Specifically , we plotted the number of binding events predicted by each algorithm at different distances between the joint binding events in Figure 2B ., Since there are two true binding events in each region , two predictions at every distance correspond to perfect positive predictive value ., dPeak on average generates more than one prediction and does not over-estimate the number of binding events when the distance between joint events is less than the average library size ., This result confirms that the higher sensitivity of dPeak in Figure 2A is not due to increased number of predictions ., In contrast , PICS and GPS on average generate only one prediction for closely located binding events , which recapitulates the conclusions from Figure 2A ., In summary , dPeak outperforms state-of-the-art deconvolution methods across different distances between joint binding events , especially when the distance between the binding events is less than the average library size ., Next , we evaluated the sensitivity and positive predictive value of the three methods on 20 , 000 candidate regions with a single binding event using the additional synthetic data from 22 ( Table S4 in Text S1 ) ., Average number of predictions per region with at least one predicted binding event and the corresponding standard errors are as follows: dPeak ( ) , PICS ( ) , GPS ( ) ., Overall , dPeak slightly over-estimates the number of binding events for regions with a single binding event , and hence PICS is slightly better than dPeak in positive predictive value for these regions ., However , as revealed by our joint event analysis , this conservative approach of PICS severely under-estimates the number of binding events when multiple events reside closely ., In contrast , GPS significantly under-estimates the number of binding events for the regions with a single binding event since it filters out too many predictions and does not result in a prediction for of the regions ., In addition , it over-estimates the number of binding events across regions for which it produces at least one prediction ., Comparisons in these two scenarios with and without joint binding events indicate that dPeak strikes a good balance between sensitivity and positive predictive value for both cases ., Once we developed dPeak as a high resolution peak detection method for both SET and PET data , we implemented simulation studies to evaluate the PET and SET assays for resolving closely spaced binding events in an unbiased manner ., Although SIPeS 24 supports PET ChIP-Seq data , we excluded it from the comparison of PET and SET ChIP-Seq datasets due to its poor performance ( Section 16 of Text S1 ) ., We generated simulated PET and SET ChIP-Seq data with two closely spaced binding events and evaluated the predictions of these two data types with dPeak ( Section 11 of Text S1; Figure S7 in Text S1 ) ., Figure 2C plots the sensitivity of dPeak as a function of distance between the joint binding events and number of reads for both the PET and SET settings ., Note that we evaluated sensitivity up to the distance of because we used windows to determine whether a binding event is correctly identified and as a result , results for the distance less than could be misleading ., When the distance between the events is at least as large as the average library size ( ) , the sensitivity using PET and SET data are comparable ., However , as the distance between joint binding events decreases , the sensitivity using SET data decreases significantly ., In contrast , PET ChIP-Seq retains its high sensitivity even for binding events that are located as close as ., As the number of reads decreases , sensitivity for both PET and SET data decreases ., When there are only DNA fragments ( i . e . , reads ) per binding event , sensitivity for PET data also decreases as the distance between joint binding events decreases ., However , even in this case , sensitivity of PET data is still significantly higher than that of SET data with much higher number of reads ., Figure 2D displays the number of binding events predicted by dPeak at different distances between joint binding events when reads correspond to each binding event for both PET and SET data and evaluates positive predictive value ., Results are similar for higher number of reads ( data not shown ) ., With PET ChIP-Seq , dPeak accurately chooses the number of binding events by BIC out of a maximum of five binding events at any distance between the joint binding events ., In contrast , SET ChIP-Seq predicts less than two binding events when the distance between the events is less than ., We present additional simulation results in Section 10 of ( Figure S6 in Text S1 ) ., These simulations reveal that even for cases with single binding events , PET has a slight advantage over SET because it predicts the location of the binding event more accurately ., Specifically , PET data always provides higher resolution compared to SET data regardless of the strength of the binding event , which we measure by the number of DNA fragments associated with the event ., For example , for a binding event with DNA fragments , the average distance between the predicted and true binding events is with a standard deviation of in the PET data whereas it is with a standard deviation of in the SET data ., Note that although this simulation procedure is based on the assumptions of dPeak model for PET data , our exploratory analysis and goodness of fit ( Figure S3A in Text S1 ) show that these assumptions hold well in the real PET ChIP-Seq data and therefore , these results have significant practical implications for real ChIP-Seq data ., Lower sensitivity of the SET compared to PET data is mainly driven by the loss of information due to unknown library size ., We describe this information loss by two concepts named invasion and truncation ( Figure 3A ) ., Top diagram of Figure 3A depicts two closely spaced binding events and a DNA fragment that is informative for the first binding event ( in red ) in the PET data ., Invasion refers to over-estimation of the library size and extension of the read to a length longer than the true one ., Equivalently , in the shifting procedure , this corresponds to shifting the read more than necessary ., As a result , the read extended to the estimated library size covers both of the closely spaced binding events in the SET data and becomes uninformative or less informative for the binding event it corresponds to ., Bottom diagram of Figure 3A also depicts two closely spaced binding events and illustrates truncation which we define as under-estimation of the library size ., In this case , the displayed DNA fragment is long and spans both binding events ( in red ) ., Therefore , it contributes to estimation of both binding events in the PET data ., In contrast , the read extended to estimated library size only covers the first binding event in the SET data and , as a result , its contribution to the first binding event is overestimated whereas its contribution to the second binding event is underestimated ., We evaluated the frequency by which fragments with invasion and truncation arise in SET data with a simulation study ., Our results ( Table S5 in Text S1 ) indicate that as high as and of the fragments for a typical peak region can be subject to invasion and truncation with the SET assay ., Figures 3B , C display the probabilities of invasion and truncation , respectively , of a DNA fragment as a function of the distance between binding events and the variance of the library size ., The analytical calculations are based on the dPeak generative model ( Section 12 of Text S1 ) ., Probabilities of invasion and truncation are higher for closely spaced binding events , especially when the library size is shorter than the estimated library size ( in this case ) ., In Figure 3B , the probability of invasion decreases for very closely spaced binding events , i . e . , when the distance between two binding events is less than ., As the distance between the binding events decreases , most DNA fragments cover both binding events and the configuration in the first diagram of Figure 3A is unlikely to occur ., Hence , there is already insufficient information to predict two binding events even in PET data and relative loss of information ( i . e . , invasion ) in SET data is insignificant ., These concepts describe how information on binding events can be lost or distorted by the incorrect estimation of the library size in the SET data ., Analytical calculations based on the dPeak generative model show that invasion and truncation influence closely located binding events the most , especially when the library size is not tightly controlled , i . e . , exhibit large variation ( Figures 3B , C ) ., We compared the performance of PET and SET sequencing for factor under the aerobic condition by generating a ‘quasi-SET data’ by randomly sampling one of the two ends of each paired reads in PET data and comparing binding events identified from both sets ., In order to match number of reads with SET data for fair comparison , only the half number of paired reads was used to construct PET data ., Comparison with the quasi-SET data controlled for the differences in the sequencing depths of the original PET and SET samples in addition to the biological variation of the replicates ., We then evaluated the dPeak predictions from the PET and SET analyses using the factor binding site annotations in the RegulonDB database as a gold standard ., Because a significant number of promoter regions lack RegulonDB annotations , we evaluated the sensitivity based on the regions that contain at least one annotated binding site ., This corresponds to binding sites in candidate regions that MOSAiCS identified ., Of these regions , harbor only a single annotated binding event ., For the regions with more than one annotated binding event , the average distance between binding events is ., dPeak analysis of the SET data identifies only of the annotated binding events ., In contrast , analysis of PET data with dPeak detects of the annotated binding sites ., Figure 4A displays average sensitivity as a function of the average distance between annotated binding events for the regions with at least two RegulonDB annotations ., A linear line is superimposed to capture the trend for both data types ., Notably , the lower sensitivity of SET compared to PET is mainly due to closely located binding events ., We also compared prediction accuracies of the PET and SET assays for the regions that harbor a single annotated binding event ., Figure 4B displays resolutions , which we define as the minimum of distances between predicted and annotated positions of binding events , achieved by the PET and SET assays ., Median resolutions are ( IQR\u200a= ) and ( IQR\u200a= ) for PET and SET , respectively ., This result indicates that positions of binding events can be more accurately predicted with the PET assay compared to SET even for regions with a single binding event ., To further examine the accuracy of the dPeak predictions , primer extension analysis was performed to map the transcription start site for eight genes ( Figures S10–S13 in Text S1; Table S7 in Text S1 ) ., dPeak analysis of the PET ChIP-Seq data predicts two closely spaced binding sites in the upstream of each of these eight genes with the distance between predictions ranging to ., Seven of these predictions are not annotated in RegulonDB and thus represent potential novel transcription start sites ., A transcription start site was detected within of ( ) of these binding site predictions ( Figure 5A and Table 1 ) , further supporting the accuracy of the dPeak PET predictions ., We treated these validated sites as a gold standard and evaluated the performance of each deconvolution algorithm for these regions ., Figure 5B depicts that dPeak with PET ChIP-Seq data attains significantly higher resolution compared to SET-based analysis regardless of the deconvolution algorithm used ( p-values of paired t-tests between dPeak using PET data and each of the other methods using SET data are ) ., dPeak with SET ChIP-Seq data has a resolution comparable to or better than those of the competing algorithms ., GPS is not included in this plot because it provides significantly worse resolution compared to other methods ( Figure S9C in Text S1 ) ., Genome-wide comparisons using the RegulonDB transcription start site annotations as a gold standard also lead to a similar conclusion , supporting the notion that PET-analysis with dPeak provides the best resolution ( Figures S9A , B in Text S1 ) ., Figures 4C and 4D display two representative peak regions from these analyses ., Figure 4C illustrates two binding events in the promoter regions of sibD and sibE genes separated by ., In this case , two peaks are easily distinguishable just by visual inspection and the predictions using both PET and SET data are comparably accurate ., Note that although these two binding events are visually distinguishable , standard applications of MACS and MOSAiCS identify this region as a single peak ., Widths of MOSAiCS and MACS peaks for this region are and , respectively ., MACS identifies the position of the right binding event as the “summit” of this region ( position ) ., Figure 4D displays the promoter region of yejG gene , where the distance between the two experimentally validated binding events is only ., In this case , dPeak application to PET data correctly predicts the number of binding events as two and identifies the locations of these events within of the validated sites ., In contrast , all of the SET-based analyses with the deconvolution algorithms ( PICS , GPS , GEM ) incorrectly predict one binding event located in the middle of the two experimentally validated binding sites ., High resolution identification of binding sites is especially important for differential occupancy analysis where a protein of interest is profiled under different conditions ., Given the high agreement between the dPeak algorithm and experimentally validated transcription start sites at a subset of promoter regions , we set out to identify differential promoter usage between the aerobic and anaerobic growth conditions by profiling the E . coli factor ., Results from the dPeak analysis of the aerobic and anaerobic PET data are summarized in Figure 5C both in the region ( i . e . , peak ) and binding event levels ., We identified peaks and dPeak binding events that were common between the and conditions ., Interestingly , only peaks were unique to the condition but dPeak analysis identified -specific binding events ., Similarly , we identified peaks unique to the condition while dPeak analysis resulted in -specific binding events ., We used the SET ChIP-Seq data from additional biological replicates under both conditions as independent validation of the results ., This independent validation using SET data identified of the binding events identified by dPeak using PET ChIP-Seq data ( of the common events , of the -specific binding events and of the -specific binding events ) ., Table S8 in Text S1 further summarizes these results by cross-tabulating the number of predicted binding events in each peak across the two conditions ., It illustrates that there are indeed many peaks with at least one binding event in each condition and different number of binding events across the two conditions ., Figure S14 in displays an example of closely located binding sites that are differentially occupied between aerobic and anaerobic conditions in PET ChIP-Seq data ., These results suggest that dPeak analysis identified many unique binding events that could not be differentiated in the peak-level analysis ., High resolution identification of binding sites with ChIP-Seq has profound effects for studying protein-DNA interactions in prokaryotic genomes and differential occupancy ., We evaluated PET and SET ChIP-Seq assays and illustrated that PET has considerably more power for deciphering locations of closely spaced binding events ., Our data-driven computational experiments indicate that when the distance between binding events gets smaller than the average library size , SET analysis have notably less power than the PET analysis ., Furthermore , PET provides better resolution than SET even when a region harbors a single binding event ., We developed and evaluated the dPeak algorithm , a model-based approach to identify protein binding site

Introduction, Results, Discussion, Materials and Methods

Chromatin immunoprecipitation followed by high throughput sequencing ( ChIP-Seq ) has been successfully used for genome-wide profiling of transcription factor binding sites , histone modifications , and nucleosome occupancy in many model organisms and humans ., Because the compact genomes of prokaryotes harbor many binding sites separated by only few base pairs , applications of ChIP-Seq in this domain have not reached their full potential ., Applications in prokaryotic genomes are further hampered by the fact that well studied data analysis methods for ChIP-Seq do not result in a resolution required for deciphering the locations of nearby binding events ., We generated single-end tag ( SET ) and paired-end tag ( PET ) ChIP-Seq data for factor in Escherichia coli ( E . coli ) ., Direct comparison of these datasets revealed that although PET assay enables higher resolution identification of binding events , standard ChIP-Seq analysis methods are not equipped to utilize PET-specific features of the data ., To address this problem , we developed dPeak as a high resolution binding site identification ( deconvolution ) algorithm ., dPeak implements a probabilistic model that accurately describes ChIP-Seq data generation process for both the SET and PET assays ., For SET data , dPeak outperforms or performs comparably to the state-of-the-art high-resolution ChIP-Seq peak deconvolution algorithms such as PICS , GPS , and GEM ., When coupled with PET data , dPeak significantly outperforms SET-based analysis with any of the current state-of-the-art methods ., Experimental validations of a subset of dPeak predictions from PET ChIP-Seq data indicate that dPeak can estimate locations of binding events with as high as to resolution ., Applications of dPeak to ChIP-Seq data in E . coli under aerobic and anaerobic conditions reveal closely located promoters that are differentially occupied and further illustrate the importance of high resolution analysis of ChIP-Seq data .

Chromatin immunoprecipitation followed by high throughput sequencing ( ChIP-Seq ) is widely used for studying in vivo protein-DNA interactions genome-wide ., Current state-of-the-art ChIP-Seq protocols utilize single-end tag ( SET ) assay which only sequences ends of DNA fragments in the library ., Although paired-end tag ( PET ) sequencing is routinely used in other applications of next generation sequencing , it has not been much adapted to ChIP-Seq ., We illustrate both experimentally and computationally that PET sequencing significantly improves the resolution of ChIP-Seq experiments and enables ChIP-Seq applications in compact genomes like Escherichia coli ( E . coli ) ., To enable efficient identification using PET ChIP-Seq data , we develop dPeak as a high resolution binding site identification algorithm ., dPeak implements probabilistic models for both SET and PET data and facilitates efficient analysis of both data types ., Applications of dPeak to deeply sequenced E . coli PET and SET ChIP-Seq data establish significantly better resolution of PET compared to SET sequencing .

journal.pcbi.1000719

Specialization Can Drive the Evolution of Modularity

For our study we consider a network of genes ., Each genes activity state is regulated by other genes in the network ., The genotype of an individual is defined as the set of the interactions among its genes ., We represent this set of interactions as a matrix ., Non-zero elements in indicate activation ( ) or repression ( ) of gene exerted by gene ., The state of the network at time is described by a vector ., A certain gene at time can be either active ( ) or inactive ( ) ., We model the change in the activity of the genes in the network according to the difference equation ( 1 ) where equals if , and it equals in all other cases ., Despite its simplicity , variants of this model have been successfully used to study how robustness can evolve in gene regulatory networks 22–24 , how robustness can aid in evolutionary innovation 25 , 26 , and how recombination can produce negative epistasis 27 ., Moreover , similar models have been successfully used to predict the dynamics of developmental processes in plants and animals 28 , 29 ., For our purpose , we consider that a phenotypic trait is defined by an attractor , a stable gene activity pattern resulting from the dynamics of a gene regulatory network ., Attractors are often associated with developmental end-states and ‘outputs’ of developmental mechanisms 22 , 30–32 ., In order to study the evolution of modularity in gene regulatory networks , we implemented evolutionary simulations that consisted of iterative rounds of mutation and selection in populations of networks ., In these simulations , we compared a set of reference gene activity patterns to actual network attractors , so that networks with attractors that were similar to the selected activity patterns had higher fitness than others ( see Methods ) ., To quantify the modularity of networks in our model , we used an algorithm 33 that identifies modules as non-overlapping densely connected groups of nodes with sparser connections between groups ( see Methods ) ., Thus , if genes in individual modules interact with many genes outside their module , the autonomy of the modules decreases , which would be reflected in a lowered modularity score ., To find out whether specialization can increase modularity , we studied 200 independent evolving populations of gene regulatory networks ( eq . 1 ) ., Each of these populations was started with identical networks , and was subject to 500 generation cycles of mutations and selection towards attainment of a fixed-point attractor I ( see Methods for details ) ., The number of generations was chosen to ensure that networks that stably attain I can arise in the population ., After gene activity pattern I had evolved , we allowed the population to evolve for 1500 more generations , but selecting for attainment of gene activity pattern I and a new pattern II during this time ., Under this selection regime , the fittest networks were those capable of stably attaining I and II from different initial conditions that may occur in different parts of a multicellular organism ., In other words , selection maintained the ability to attain I while at the same time favoring acquisition of II ., Pattern II was chosen such that half of the network genes had identical ( shared ) expression states in I and II , and the other half differed in their activity state ( Figure 1A ) ., We chose such activity patterns because we hypothesized that interactions between genes with shared activity states and the rest of the genes would obstruct either, i ) the constant activity state of the former , or, ii ) the capacity of the latter to acquire different activity states independently of genes with constant activity states ., If so , interactions between the different sets of genes may be selected against , thus resulting in two sets of genes with only sparse connections between them ., In most of the 200 evolving populations , modularity increased after evolving towards the attainment of both I and II ., We observe this increase both in the networks with the highest fitness in the population ( Figure 1B; Wilcoxon signed-rank test; ; ) , and when averaged over all networks in a population ( Figure 1C; Wilcoxon signed-rank test; ; ) ., Figures 1D , E show an example of how modularity increases after selection for attainment of activity patterns I and II ., Modularity does not increase when selection for II is absent , nor when networks evolve in the absence of selection ( Figure S2 ) ., The increase in modularity is not transient because it is maintained around the same level , at least for 10 , 000 additional generations , when selecting for both I and II ( Figure 2 ) ., We next verified that our results were insensitive to changes in model assumptions and parameters ., We first decreased the mutation rate , and even though the time required to evolve activity patterns I and II then increases , modularity still increases significantly ( ; Figure S3A ) ., Modularity increases as well when is increased ( ; Figure S3B ) ., We next asked whether our observations were sensitive to the assumption that individual gene activity patterns contribute to fitness additively ., Changing this assumption to multiplicative fitness contributions still leads to a significant increase in modularity ( Figure S4 ) ., In addition , the increase in modularity also occurs for networks containing more genes ( ; Figure S5 ) , suggesting that such behavior does not depend on the number of genes in a network ., In a next analysis , we asked whether the increase in modularity depends on the identity of gene activity states I and II ., We found that it does not , as long as some genes have the same activity state in the two patterns ., For example , modularity also increases when the activity patterns differ in the activity of either three or seven genes ( Figure S6A , B ) ., Moreover , modularity increases when both the first and the second gene activity patterns are randomly chosen , except that pairs with fewer than two different activity states are discarded ( Figure S6C , based on 100 populations with different pairs of activity patterns ) ., In contrast , modularity does not increase when all genes in the activity states I and II differ in their expression ( Figure S6D ) ., This result is not due to a lack of adaptation , since networks able to attain both activity patterns arise in all evolving populations ., Taken together , these observations show that modularity does not only increase for specific gene activity patterns , but that it is a generic evolutionary response ., Moreover , the distinction between two sets of genes , those with identical and those with different activity in both expression patterns , is essential for the evolution of modularity ., That modularity increases only in this case suggests that modules arise as a means of diminishing the effects of genes with unchanging activity on genes with changing expression in I and II , and vice versa ., If so , modules should correspond to sets of genes that are required to switch their activity in a concerted manner ., The following section shows that this is the case ., Having established that the evolution of modularity requires genes with both shared and different activity states , we next asked whether the partitioning of modules is congruent with these two sets of genes ., In other words , does one module tend to involve the genes with shared activity states , whereas another involves genes with different activity states in I and II ?, We evolved 300 network populations , first towards activity pattern I and later towards both I and II , depicted in Figure 1A ., Throughout evolution , we determined for one of the best adapted networks in each population:, i ) the frequency at which two genes with activity states shared in I and II occur within the same module ,, ii ) the frequency at which two genes with different activity states in I and II occur within the same module ,, iii ) the frequency with which a specific gene with a shared activity state and a gene with a non-shared activity state are in the same module ( Figure 3A ) ., As selection for I and II occurs , and increase , while decreases ( Figure 3B ) ., This observation tells us that genes with activity states that change concertedly throughout all the selected activity patterns – be they shared or not – will tend to be included in the same module , and kept apart from other genes ., This is exemplified in Figure 1D , E , which compares one of the optimal networks after selection for I with one of the optimal networks after selection for both I and II ., The latter is partitioned into modules in which genes with shared and distinct activity states in I and II lie apart ., Thus , the structure of modules reflects the manner in which selection has molded the traits , as has been previously suggested 2 ., We also tested whether modularity arises only where selection favors the attainment of two gene activity patterns , or whether it increases further with even more gene activity patterns ., To this end , we analyzed 100 evolving populations in which selection first favored a gene activity pattern I ( 500 generations ) , then an additional pattern II ( I+II , next 1 , 500 generations ) , and then a third pattern III ( I+II+III , last 3 , 000 generations ) ., The patterns share the activity of some genes and differ in others ., As selection for the third pattern begins , more and smaller groups of genes arise whose activity changes in a concerted manner ( Figure 4A ) ., Interactions between different such groups would obstruct evolutionary adaptation ., Such interactions should thus be selected against , resulting in a further increase in modularity ., Our observations confirm this hypothesis ., After selection for patterns I and II , we observed a significant first increase in modularity ( Wilcoxon signed-rank test; ; ) ., Modularity increased further after selection for pattern III ( Figure 4B; Wilcoxon signed-rank test; ; ) ., In addition , we observed an increased number of modules in networks with high fitness after selection for patterns I and II ., Moreover , this number increases further after selection for patterns I , II and III ( Figure S8B ) ., This result suggests that the increase in modularity after selection for the three patterns occurs because of the appearance of new modules , and is not a mere consequence of the consolidation and refinement of previously evolved modules ., We also analyzed how the probability of two genes being part of the same module changes across evolution ., We found that the frequency of two genes occurring in the same module in the fittest networks of each evolving population changes according to whether those genes change their activity concertedly across the selected patterns ( Figure S8C ) ., For example , as we depict in Figures 4 and S8A , the activity of genes 5 and 6 changes concertedly across all activity patterns: if in one pattern gene 5 is active , then gene 6 is inactive in that same pattern , and vice versa ., The frequency with which those genes lie in the same module increases across evolution ., In contrast , the activity of genes 0 and 6 changes concertedly when selecting for patterns I and II , but not when also selecting for activity pattern III ., Thus , the probability of those genes occurring in the same module increases prior to selection for pattern III ., After selection for pattern III starts , the probability that genes 0 and 6 lie in the same module decreases abruptly ( Figure S8C ) ., These results show that the modules that arise after selection for the third pattern also tend to coincide with sets of genes whose activity states change concertedly throughout the selected patterns ., Computational cost did not allow exploration of further increases in modularity via selection of additional gene activity patterns ., However , our observations already suggest that modularity will increase as long as there is an increase in the number of gene groups for which concerted activity changes are favored ., A question recurring in the literature is how modularity may increase evolvability by facilitating co-option , the combination of previously evolved modules to perform new functions 19–21 , 34–36 ., We addressed how the previous evolution of modules in gene regulatory networks biases future evolutionary potential by asking whether gene networks acquire new gene activity patterns faster if these patterns use gene activity states associated with previously evolved modules ., Specifically , we selected networks for their ability to stably attain three gene activity patterns I , II and III ( Figure 5A ) ., We chose the specific combination of patterns in Figure 5A because:, i ) it promotes the evolution of a module including genes 0–4 and another module including genes 5–9 , as shown above , and ,, ii ) it allows the inclusion of an additional activity pattern ( IV ) that is composed entirely of activity states associated with previously evolved modules ( Figure 5A , B ) ., After 3 , 000 generations , we subjected networks in 100 evolving populations to selection favoring such an additional gene activity pattern IV ( Figure 5B ) ., Importantly , this pattern shares the activity states of genes 0–4 with III , and the activity state of genes 5–9 with II ., Thus , gene activity pattern IV may evolve by combining previously evolved modules in a new manner ., In addition , we repeated this approach in 100 “control” populations where the fourth favored gene activity pattern was randomly chosen with equal probability for genes being active and inactive ., Notice that we do not expect that selection for activity pattern IV increases modularity , because the inclusion of this pattern does not cause an increase in the number of gene groups with concerted activity changes ., Rather , we hypothesize that modularity facilitates the evolutionary acquisition of such an activity pattern , as compared to other activity patterns ., We found that networks with high fitness arise much more rapidly when IV is the new gene activity pattern ., This indicates that pattern IV is much easier to attain than random gene activity patterns in populations of networks that have previously been selected for their ability to attain I , II and III ( Figure 5C ) ., The same trend occurs when not just the networks with highest fitness are considered , but also when we analyze mean population fitness ( Figure S7 ) ., We note that in our analysis selection favors the attainment of IV to the same extent as the attainment of any one random gene activity pattern in the control populations ., This means that our observations are not simply caused by a greater increase in fitness conveyed by IV ., The fitness increase rather depends on how easily the new gene activity patterns can be constructed: it is easier to evolve gene activity patterns that combine activity states of previously evolved modules ., In sum , we showed here that modularity arises in gene networks when they acquire the ability to attain new activity patterns that share the activity state of some genes with old patterns ., Our observations indicate that selection to attain the new activity patterns can cause modularity to arise in gene regulatory networks when pleiotropic effects obstruct adaptation 2 , 8 , 11 ., Such pleiotropic effects are caused by interactions between, ( i ) genes whose activity is shared between different patterns , and, ( ii ) genes whose activity is specific to one pattern: If changes in the latter affect the former , evolutionary acquisition of the new pattern is hindered ., Thus , the scenario we propose favors networks with few interactions between genes with an unchanging activity state and genes that adopt new regulatory functions ., In this way , genes that have correlated activity states come to lie in the same network module ( Figures 3 and S8C ) ., Our results suggest that modularity increases as long as selection favors new activity patterns involving more and smaller groups of genes whose activity changes in a concerted manner ( Figures 4 and S8 ) ., Empirical falsification ( or validation ) of the mechanism that we propose ideally requires comparative analyses of the structure of gene regulatory networks in several related species ., Such information might not be available soon ., However , existing information from various sources suggests that the mechanism we propose could be important ., Specifically , the evolutionary acquisition of new gene activity states by regulatory networks is ubiquitous in evolution , and nowhere more than in the evolution of development ., It occurs wherever new cell types , organs , or body structures , arise from previously undifferentiated ones ., Many examples in the literature suggest that some genes exhibit specialized activity in different parts of an organism , whereas others present shared activity patterns ., Indeed , gene functions may be inferred via correlated gene expression patterns in conventional or high-throughput expression analyses 37–39 ., For example , the activity of the same genes patterns both vegetative and floral meristems in the plant Arabidopsis thaliana ., Floral identity genes are active exclusively in floral meristems , so that the floral structure is determined by both the floral identity genes and the shared patterning genes 40–42 ., In the sea urchin Strongylocentrus purpuratus , some differentiation genes are active in the micromer lineage that produces the euechinoid exclusive embryonic skeleton and also in the independently derived juvenile skeletogenic centers that produce the adult skeleton 43 ., Some other genes of the gene network that specifies the skeletogenic micromere lineage are active in those cells but not in the juvenile skeletogenic centers ., Examples include genes involved in induction of neighbouring cells or in triggering the initial stages of micromere specification 43 , 44 ., Another example involves the cellular level ., Mammalian brown fat cells share some traits and gene activity patterns with white fat cells , and others with muscle cells 45 ., More generally , evolutionarily derived cell types usually perform just a fraction of the functions that ancestral cell types performed 46 , a trend that will lead to similar activity states for some genes and different states for others in sister cell types ., In a similar vein , evolutionary specialization of initially homogeneous metameric units is likely to occur mainly by modifications ( such as changes in the transcriptional circuitry ) that result in metamers with different activity states of some genes but not of others; otherwise , differentiated metameric units would be hardly recognizable as such ., For example , in D . melanogaster , limbs are positioned and patterned by mechanisms that are reiterated along the body , however limb identity relies on segment-specific mechanisms 47 ., Moreover , in heteronomous arthropods , in which the morphology of segments along an individual is very distinct , processes underlying segmentation and limb differentiation interact less than in homonomous arthropods , in which the segments along a body are very similar 47 ., Segment formation is performed throughout the organism ( shared ) , and , in heteronomous taxa , limb identity determination is specialized according to the place where a limb develops ., Thus , when there is specialization in limb identity , the two processes are more independent , in contrast to taxa that lack this specialization ., Co-option , the recruitment of previously evolved modules to perform new functions , is a common feature of evolutionary innovations 20 , 21 , 34 , 36 , 43 ., A case in point regards the gene network regulating pharyngeal dentition in fish , which is co-opted to also generate oral dentition 36 ., Another example is the gene network that patterns the insect wing blade ., It is co-opted to determine the localization of eyespots in butterfly wings 34 ., Our work shows that a modular network may readily generate new gene activity patterns that make use of gene activity states of previously evolved modules ., The existence of such structured , or “facilitated” variation has been known for a long time 48–51 ., Our work provides a candidate mechanism to create such variation , namely via network modularity that results from specialization in gene activity ., Our observations could thus help explain the repetitive co-option of several modules , such as that responsible for proximal-distal polarity in lateral appendages and body outgrowths 20 , 21 , or the achaete and scute module that operates in a wide range of developmental processes in animals 19 ., An alternative hypothesis for the evolution of modularity is the ‘modularly-varying goals’ scenario 10 ., This scenario requires that populations are exposed to evolutionary goals that fluctuate over time , so that modularity can arise and be maintained ., In contrast , our scenario requires specialization of gene activity , that is , new gene activity patterns must be attained while old activity patterns are preserved ., Relatedly , the modularly-varying goals scenario requires genetic changes for evolutionary adaptation after an evolutionary goal changes ., In contrast , our mechanism requires one genotype to produce different activity patterns under different conditions , conditions that may occur in different parts of a multicellular organism ., In other words , in our scenario , modularity arises to avoid obstruction to attain different selected patterns within the same genotype ., Our scenario may thus be more appropriate for traits where environmental demands are not constantly fluctuating , such as in the development of many morphological traits in plants and animals ., Thus far , we motivated our approach with the development of multicellular organisms ., However , the approach could also explain modularity in unicellular organisms ., For example , the metabolic networks of bacteria living in changing environments tend to be more modular than those of bacteria living in stable environments 12 ., Similar patterns may exist for gene regulatory networks ., If so , the modularly varying goals scenario is not their only possible explanation ., Unicellular organisms respond to changing environments by tuning their gene activity pattern ., In other words , they usually have adaptively plastic phenotypes ., For example , different sets of genes are activated or repressed when yeast cells are exposed to different environments 52–54 ., Evolving the ability to switch gene expression according to the environment requires producing several alternative activity patterns , as we propose here ., Importantly , some yeast genes change their expression concertedly in several environments , whereas others have responses that are specific to any one environment 52–54 ., This observation suggests that the activity of some genes is shared across alternative activity patterns while the activity of other genes is particular to certain environments , as our model demands ., In sum , because organisms in changing environments are required to produce different gene activity patterns according to the environment , our scenario can explain the evolution of modularity both in fluctuating and non-fluctuating environments ., A question that remains unanswered is whether our model applies to genotype-phenotype maps different from those of gene regulatory networks ., A prominent example is metabolic networks , whose phenotypes are patterns of metabolic fluxes through network reactions ., Our framework may apply to some instances of modularity in metabolic systems , as the following example illustrates ., The main requirement of our model is an increase in the number of functions that a network must perform ( i . e . in the number of selected gene activity patterns ) ., The appearance of new functions in a metabolic network usually involves the production of new metabolites ., Hintze and Adami 55 performed evolutionary simulations of an artificial metabolism in which the fittest metabolic networks were able to produce an increasingly diverse spectrum of metabolites ., This selection regime resulted in increased modularity of metabolic networks , an observation consistent with the mechanism that we propose for gene regulatory networks ., Our work aimed at conceptual clarity by using only few essential assumptions in explaining the evolution of modularity ., We therefore neglected many processes that doubtlessly play a major role in the evolution of regulatory gene networks ., For example , we did not consider mutations changing the number of genes in a network , even though processes such as gene loss or duplication may be frequently involved in the appearance of new gene activity patterns ., Similarly , the appearance of new body structures or cell types requires interactions among cells , tissues and organs ., Such interactions ensure the proper placement of cells with the combination of general and specialized gene activity that is characteristic of specialization ., The incorporation of these and other processes in future work will deepen our understanding of the evolution of modularity , and thus of evolvability ., We here identify modularity using one 33 , 56 of several algorithms aimed at identifying structural modules , densely connected groups of nodes with sparser connections between groups ., The measure of modularity in this algorithm is a score that compares the abundance of intra-module connections between a given network to that of random networks with the same degree distribution 57 ., is defined as: ( 2 ) where denotes one of the prospective modules in a network , stands for the total number of edges in the network , represents the number of edges within module , and is the sum of the number of connections that each node in module has 33 , 56–58 ., The algorithm we use 33 identifies a partitioning of networks into modules that maximizes ., We use this algorithm because of its computational efficiency and accuracy 33 , 56 ., We also explored different algorithms 57 , 58 and found that our results hold regardless of these choices ., Typical values of partitions that maximize intra-module connections in random networks vary depending on the number of nodes , edges and connectivity distribution 59 ., For example , the maximum value of a network varies as a function of the total number of edges in it 60 ., Hence , a fair comparison of modularity in different networks requires first addressing how atypical is in the best partition of each network when compared with random networks with the same attributes ., Following 10 we use for normalization the equation: ( 3 ) where is the modularity returned by the Newman algorithm 33 , 56 for a certain network , stands for the average value of 1 , 000 random networks with the same number of genes and edges and the same degree distribution as the original network ., values for these random networks are also calculated using the Newman algorithm ., is the maximal value in these 1 , 000 random networks ., The normalized modularity tells us how modular a network is in comparison to random networks with the same attributes ., Non-normalized and normalized values render equivalent results in our analysis ( Figure S1 ) ., Therefore , we restrict ourselves to report results for normalized modularity ., The fitness function we use compares a set of reference gene activity patterns to actual network attractors ., Our fitness measure also incorporates the likelihood that an attractor is attained in the face of perturbations ., In doing so , it takes into account not only the identity of an attractor but also its robustness , an important feature for the stability and reproducibility of developmental processes 13 , 18 , 61–63 ., For each gene activity pattern that contributes to fitness and for each network in our analysis , evaluation of fitness involved the following steps:, i ) The initial state of the gene network at time 0 was chosen to be a perturbation of the target pattern , drawn from a probability distribution where the initial state of each gene differs from that of with probability ., ii ) We carried out network dynamics ( eq . 1 ) until some new attractor was reached;, iii ) We recorded the Hamming distance ( ) separating from , and calculated the contribution to fitness of this developmental trajectory as ; modifications of by varying the exponent produce equivalent results ., iv ) We repeated steps, i ) –iii ) 500 times to determine 500 values ( ) ., Notice that several of such 500 values would correspond to the same initial condition , and that the distribution of possible initial conditions is biased towards gene activity patterns similar to the reference pattern ., This reflects our assumption , for the sake of simplicity , that selection favors similar initial conditions leading to the same selected activity pattern ., We also assumed that gene activity patterns that are similar to the reference pattern are more likely to be required as initial conditions ., Relaxation of such assumptions by variation in did not modify our results ., We then calculated the networks fitness as ( 4 ) where is the arithmetic mean of all ., Wherever fitness needed to be evaluated for multiple gene activity patterns , we calculated the arithmetic mean of over these multiple patterns ., Notice that selection is pushing the acquisition of different gene activity patterns that would appear under different conditions ( such as different parts of the organism ) ., Hence , the optimal networks will be those with dynamics that lead to different attractors matching the reference activity patterns , and not those with a single attractor that is a combination of the reference patterns ., Had we used multiplicative contributions to fitness then the benefits that result from attaining a gene activity pattern would have depended on the acquisition of all other activity patterns ., Because our simulations start with selection for a single activity pattern , it was preferable to assume otherwise ., Using additive contributions to fitness guarantees that networks that are not able to attain the new gene activity pattern still have a chance to contribute to the next generation ., However , usage of multiplicative fitness contributions does not affect our results qualitatively ., For each simulation of gene network evolution , we first built a 10 node network and added 20 interactions at random to its interaction matrix ., These interactions were activating or repressing , with equal probability ., To construct the initial population we exposed 100 copies of this initial network to random mutation ., Mutations occurred independently among different genes ., A mutation of a gene either added a positive or negative interaction affecting the genes activity , or eliminated one of the interactions that regulated the gene ., Such mutations can be interpreted as changes in the regulatory regions of a gene , adding or eliminating cis-regulatory elements ., Most of our results are based on a probability of a mutation occurring in a gene ( ) of 0 . 05 ., This value of allowed adaptation within a tractable number of generations ., Variation in did not affect our results , but only affected the time required for adaptation ., For a gene undergoing mutation , we defined the probability of losing an interaction as ( 5 ) and the probability of acquiring a new interaction as ., Here represents the number of regulators of the gene , and equals the number of genes in the network , and hence , the maximum number of regulators of any gene ., This procedure results in networks that evolve towards low connectivities of 2–3 regulators per gene ., Such low connectivity is often observed in transcriptional regulation networks of plants , animals , fungi and bacteria 64 ., Loss of interactions may also help explain the observation that loss of gene expression is more common than acquiring new expre

Introduction, Results, Discussion, Methods

Organismal development and many cell biological processes are organized in a modular fashion , where regulatory molecules form groups with many interactions within a group and few interactions between groups ., Thus , the activity of elements within a module depends little on elements outside of it ., Modularity facilitates the production of heritable variation and of evolutionary innovations ., There is no consensus on how modularity might evolve , especially for modules in development ., We show that modularity can increase in gene regulatory networks as a byproduct of specialization in gene activity ., Such specialization occurs after gene regulatory networks are selected to produce new gene activity patterns that appear in a specific body structure or under a specific environmental condition ., Modules that arise after specialization in gene activity comprise genes that show concerted changes in gene activities ., This and other observations suggest that modularity evolves because it decreases interference between different groups of genes ., Our work can explain the appearance and maintenance of modularity through a mechanism that is not contingent on environmental change ., We also show how modularity can facilitate co-option , the utilization of existing gene activity to build new gene activity patterns , a frequent feature of evolutionary innovations .

Throughout lifes history , organisms have produced evolutionary innovations , features that are useful when facing new ecological and environmental challenges ., A property that aids in the production of such innovations is modularity ., Modular systems consist of groups of molecules with many interactions within a group but fewer interactions between groups ., Such modularity increases the chances of innovation , because it allows changes inside one module without perturbing others , and because it permits redeployment of modules to create new biological functions ., We simulate the evolution of gene networks known to be important in development to show that modularity increases when selection favors specialization in gene activity ., Specialization occurs wherever new cell types , organs , or other body structures arise ., In the course of this process gene networks acquire the ability to produce new gene activity patterns specific to these structures ., We also demonstrate how modularity favors the evolution of new gene activity patterns that make use of already existing modules ., Because specialization in gene activity is very common in evolution , the mechanism that we put forward may be important for the origins of modularity in gene regulatory networks .

computational biology/evolutionary modeling, developmental biology/developmental evolution, evolutionary biology/developmental evolution

journal.pcbi.1003037

Bayesian Computation Emerges in Generic Cortical Microcircuits through Spike-Timing-Dependent Plasticity

Numerous experimental data show that the brain applies principles of Bayesian inference for analyzing sensory stimuli , for reasoning and for producing adequate motor outputs 1–5 ., Bayesian inference has been suggested as a mechanism for the important task of probabilistic perception 6 , in which hidden causes ( e . g . the categories of objects ) that explain noisy and potentially ambiguous sensory inputs have to be inferred ., This process requires the combination of prior beliefs about the availability of causes in the environment , and probabilistic generative models of likely sensory observations that result from any given cause ., By Bayes Theorem , the result of the inference process yields a posterior probability distribution over hidden causes that is computed by multiplying the prior probability with the likelihood of the sensory evidence for all possible causes ., In this article we refer to the computation of posterior probabilities through a combination of probabilistic prior and likelihood models as Bayesian computation ., It has previously been shown that priors and models that encode likelihoods of external stimuli for a given cause can be represented in the parameters of neural network models 6 , 7 ., However , in spite of the existing evidence that Bayesian computation is a primary information processing step in the brain , it has remained open how networks of neurons can acquire these priors and likelihood models , and how they combine them to arrive at posterior distributions of hidden causes ., The fundamental computational units of the brain , neurons and synapses , are well characterized ., The synaptic connections are subject to various forms of plasticity , and recent experimental results have emphasized the role of STDP , which constantly modifies synaptic strengths ( weights ) in dependence of the difference between the firing times of the pre- and postsynaptic neurons ( see 8 , 9 for reviews ) ., Functional consequences of STDP can resemble those of rate-based Hebbian models 10 , but may also lead to the emergence of temporal coding 11 and rate-normalization 12 , 13 ., In addition , the excitability of neurons is modified through their firing activity 14 ., Some hints about the organization of local computations in stereotypical columns or so-called cortical microcircuits 15 arises from data about the anatomical structure of these hypothesized basis computational modules of the brain ., In particular , it has been observed that local ensembles of pyramidal neurons on layers 2/3 and layers 5/6 typically inhibit each other , via indirect synaptic connections involving inhibitory neurons 16 ., These ubiquitous network motifs were called soft winner-take-all ( WTA ) circuits , and have been suggested as neural network models for implementing functions like non-linear selection 16 , 17 , normalization 18 , selective attention 19 , decision making 20 , 21 , or as primitives for general purpose computation 22 , 23 ., A comprehensive theory that explains the emergence of computational function in WTA networks of spiking neurons through STDP has so far been lacking ., We show in this article that STDP and adaptations of neural excitability are likely to provide the fundamental components of Bayesian computation in soft WTA circuits , yielding representations of posterior distributions for hidden causes of high-dimensional spike inputs through the firing probabilities of pyramidal neurons ., This is shown in detail for a simple , but very relevant feed-forward model of Bayesian inference , in which the distribution for a single hidden cause is inferred from the afferent spike trains ., Our new theory thus describes how modules of soft WTA circuits can acquire and perform Bayesian computations to solve one of the fundamental tasks in perception , namely approximately inferring the category of an object from feed-forward input ., Neural network models that can handle Bayesian inference in general graphical models , including bi-directional inference over arbitrary sets of random variables , explaining away effects , different statistical dependency models , or inference over time require more complex network architectures 24 , 25 , and are the topic of ongoing research ., Such networks can be composed out of interconnected soft WTA circuits , which has been shown to be a powerful principle for designing neural networks that can solve arbitrary deterministic or stochastic computations 22 , 23 , 25 ., Our theory can thus be seen as a first step towards learning the desired functionality of individual modules ., At the heart of this link between Bayesian computation and network motifs of cortical microcircuits lies a new theoretical insight on the micro-scale: If the STDP-induced changes in synaptic strength depend in a particular way on the current synaptic strength , STDP approximates for each synapse exponentially fast the conditional probability that the presynaptic neuron has fired just before the postsynaptic neuron ( given that the postsynaptic neuron fires ) ., This principle suggests that synaptic weights can be understood as conditional probabilities , and the ensemble of all weights of a neuron as a generative model for high-dimensional inputs that - after learning - causes it to fire with a probability that depends on how well its current input agrees with this generative model ., The concept of a generative model is well known in theoretical neuroscience 26 , 27 , but it has so far primarily been applied in the context of an abstract non-spiking neural circuit architecture ., In the Bayesian computations that we consider in this article , internal generative models are represented implicitly through the learned values of bottom-up weights in spiking soft-WTA circuits , and inference is carried out by neurons that integrate such synaptic inputs and compete for firing in a WTA circuit ., In contrast to previous rate-based models for probabilistic inference 28–30 every spike in our model has a clear semantic interpretation: one spike indicates the instantaneous assignment of a certain value to an abstract variable represented by the firing neuron ., In a Bayesian inference context , every input spike provides evidence for an observed variable , whereas every output spike represents one stochastic sample from the posterior distribution over hidden causes encoded in the circuit ., We show that STDP is able to approximate the arguably most powerful known learning principle for creating these implicit generative models in the synaptic weights: Expectation Maximization ( EM ) ., The fact that STDP approximates EM is remarkable , since it is known from machine learning that EM can solve a fundamental chicken-and-egg problem of unsupervised learning systems 31: To detect - without a teacher - hidden causes for complex input data , and to induce separate learning agents to specialize each on one of the hidden causes ., The problem is that as long as the hidden causes are unknown to the learning system , it cannot tell the hidden units what to specialize on ., EM is an iterative process , where initial guesses of hidden causes are applied to the current input ( -step ) and successively improved ( -step ) , until a local maximum in the log-likelihood of the input data is reached ., In fact , the basic idea of EM is so widely applicable and powerful that most state-of-the art machine learning approaches for discovering salient patterns or structures in real-world data without a human supervisor rely on some form of EM 32 ., We show that in our spiking soft-WTA circuit each output spike can be viewed as an application of the -step of EM ., The subsequent modification of the synaptic weights between the presynaptic input neurons and the very neuron that has fired the postsynaptic spike according to STDP can be viewed as a move in the direction of the -step of a stochastic online EM procedure ., This procedure strives to create optimal internal models for high-dimensional spike inputs by maximizing their -likelihood ., We refer to this interpretation of the functional role of STDP in the context of spiking WTA circuits as spike-based Expectation Maximization ( SEM ) ., This analysis gives rise to a new perspective of the computational role of local WTA circuits as parts of cortical microcircuits , and the role of STDP in such circuits: The fundamental computational operations of Bayesian computation ( Bayes Theorem ) for the inference of hidden causes from bottom-up input emerge in these local circuits through plasticity ., The pyramidal neurons in the WTA circuit encode in their spikes samples from a posterior distribution over hidden causes for high-dimensional spike inputs ., Inhibition in the WTA accounts for normalization 18 , and in addition controls the rate at which samples are generated ., The necessary multiplication of likelihoods ( given by implicit generative models that are learned and encoded in their synaptic weights ) with simultaneously learned priors for hidden causes ( in our model encoded in the neuronal excitability ) , does not require any extra computational machinery ., Instead , it is automatically carried out ( on the scale ) through linear features of standard neuron models ., We demonstrate the emergent computational capability of these self-organizing modules for Bayesian computation through computer simulations ., In fact , it turns out that a resulting configuration of networks of spiking neurons can solve demanding computational tasks , such as the discovery of prototypes for handwritten digits without any supervision ., We also show that these emergent Bayesian computation modules are able to discover , and communicate through a sparse output spike code , repeating spatio-temporal patterns of input spikes ., Since such self-adaptive computing and discrimination capability on high-dimensional spatio-temporal spike patterns is not only essential for early sensory processing , but could represent a generic information processing step also in higher cortical areas , our analysis suggests to consider networks of self-organizing modules for spike-based Bayesian computation as a new model for distributed real-time information processing in the brain ., Preliminary ideas for a spike-based implementation of EM were already presented in the extended abstract 20 , where we analyzed the relationship of a simple STDP rule to a Hebbian learning rule , and sketched a proof for stochastic online EM ., In the present work we provide a rigorous mathematical analysis of the learning procedure , a proof of convergence , expand the framework towards learning spatio-temporal spike patterns , and discuss in detail the relationship of our STDP rule to experimental results , as well as the interpretation of spikes as samples from instantaneous posterior probability distributions in the context of EM ., Our model consists of a network of spiking neurons , arranged in a WTA circuit , which is one of the most frequently studied connectivity patterns ( or network motifs ) of cortical microcircuits 16 ., The input of the circuit is represented by the excitatory neurons ., This input projects to a population of excitatory neurons that are arranged in a WTA circuit ( see Fig . 1 ) ., We model the effect of lateral inhibition , which is the competition mechanism of a WTA circuit 33 , by a common inhibitory signal that is fed to all neurons and in turn depends on the activity of the neurons ., Evidence for such common local inhibitory signals for nearby neurons arises from numerous experimental results , see e . g . 16 , 34–36 ., We do not a priori impose a specific functional relationship between the common inhibition signal and the excitatory activity ., Instead we will later derive necessary conditions for this relationship , and propose a mechanism that we use for the experiments ., The individual units are modeled by a simplified Spike Response Model 37 in which the membrane potential is computed as the difference between the excitatory input and the common inhibition term ., sums up the excitatory inputs from neurons as ( 2 ) models the EPSPs evoked by spikes of the presynaptic neuron , and models the intrinsic excitability of the neuron ., In order to simplify our analysis we assume that the EPSP can be modeled as a step function with amplitude , i . e . , it takes on the value 1 in a finite time window of length after a spike and is zero before and afterwards ., Further spikes within this time window do not contribute additively to the EPSP , but only extend the time window during which the EPSP is in the high state ., We will later show how to extend our results to the case of realistically shaped and additive EPSPs ., We use a stochastic firing model for , in which the firing probability depends exponentially on the membrane potential , i . e . , ( 3 ) which is in good agreement with most experimental data 38 ., We can thus model the firing behavior of every neuron in the WTA as an independent inhomogeneous Poisson process whose instantaneous firing rate is given by ., In order to understand how this network model generates samples from a probability distribution , we first observe that the combined firing activity of the neurons in the WTA circuit is simply the sum of the independent Poisson processes , and can thus again be modeled as an inhomogeneous Poisson process with rate ., Furthermore , in any infinitesimally small time interval , the neuron spikes with probability ., Thus , if we know that at some point in time , i . e . within , one of the neurons produces an output spike , the conditional probability that this spike originated from neuron can be expressed as ( 4 ) Every single spike from the WTA circuit can thus be seen as an independent sample from the instantaneous distribution in Eq ., ( 4 ) at the time of the spike ., Although the instantaneous firing rate of every neuron directly depends on the value of the inhibition , the relative proportion of the rate to the total WTA firing rate is independent of the inhibition , because all neurons receive the same inhibition signal ., Note that determines only the value of the sample at time , but not the time point at which a sample is created ., The temporal structure of the sampling process depends only on the overall firing rate ., This implementation of a stochastic WTA circuit does not constrain in any way the kind of spike patterns that can be produced ., Every neuron fires independently according to a Poisson process , so it is perfectly possible ( and sometimes desirable ) that there are two or more neurons that fire ( quasi ) simultaneously ., This is no contradiction to the above theoretical argument of single spikes as samples ., There we assumed that there was only one spike at a time inside a time window , but since we assumed these windows to be infinitesimally small , the probability of two spikes occurring exactly at the same point in continuous time is zero ., Fig . 3 demonstrates the emergence of Bayesian computation in the generic network motif of Fig . 1A in a simple example ., Spike inputs ( top row of Fig . 3D ) are generated through four different hidden processes ( associated with four different colors ) ., Each of them is defined by a Gauss distribution over a 2D pixel array with a different center , which defines the probability of every pixel to be on ., Spike trains encode the current value of a pixel by a firing rate of 25 Hz or 0 Hz for 40 ms . Each pixel was encoded by two input neurons via population coding , exactly one of them had a firing rate of 25 Hz for each input image ., A 10 ms period without firing separates two images in order to avoid overlap of EPSPs for input spikes belonging to different input images ., After unsupervised learning with STDP for 500 s ( applied to continuous streams of spikes as in panel D of Fig . 3 ) the weight vectors shown in Fig . 3F ( projected back into the virtual 2D input space ) emerged for the four output neurons , demonstrating that these neurons had acquired internal models for the four different processes that were used to generate inputs ., The four different processes for generating the underlying 2D input patterns had been used with different prior probabilities ( ) ., Fig . 3G shows that this imbalance resulted in four different priors encoded in the biases of the neurons ., When one compares the unequal sizes of the colored areas in Fig . 3H with the completely symmetric internal models ( or likelihoods ) of the four neurons shown in panel F , one sees that their firing probability approximates a posterior over hidden causes that results from multiplying their learned likelihoods with their learned priors ., As a result , the spike output becomes sparser , and almost all neurons only fire when the current input spikes are generated by that one of the four hidden processes on which they have specialized ( Fig . 3D , bottom row ) ., In Fig . 3I the performance of the network is quantified over time by the normalized conditional entropy , where is the correct hidden cause of each input image in the training set , and denotes the discrete random variable defined by the firing probabilities of output neurons for each image under the currently learned model ., Low conditional entropy indicates that each neuron learns to fire predominantly for inputs from one class ., Fig . 3E as well as the dashed blue line in Fig . 3I show that the learning process is improved when a common background oscillation at 20 Hz is superimposed on the firing rate of input neurons and the membrane potential of the output neurons , while keeping the average input and output firing rates constant ., The reason is that in general it may occur that an output neuron receives during its integration time window ( 40 ms in this example ) no information about the value of a pixel ( because neither the neuron that has a high firing rate for 40 ms if this pixel is black , nor the associated neuron that has a high firing rate if this pixel is white fire during this time window ) ., A background oscillation reduces the percentage of such missing values by driving presynaptic firing times together ( see top row of Fig . 3E ) ., Note that through these oscillations the overall output firing rate fluctuates strongly , but since the same oscillation is used consistently for all four types of patterns , the circuit still learns the correct distribution of inputs ., This task had been chosen to become very fast unsolvable if many pixel values are missing ., Many naturally occurring input distributions , like the ones addressed in the subsequent computer experiments , tend to have more redundancy , and background oscillations did not improve the learning performance for those ., In this section we will develop the link between the unsupervised learning of the generative probabilistic model in Fig . 1B and the learning effect of STDP as defined in our spiking network model in Fig . 1A ., Starting from a learning framework derived from the concept of Expectation Maximization 31 , we show that the biologically plausible STDP rule from Fig . 2 can naturally approximate a stochastic , online version of this optimization algorithm ., We call this principle SEM ( spike-based EM ) ., SEM can be viewed as a bootstrapping procedure ., The relation between the firing probabilities of the neurons within the WTA circuit and the continuous updates of the synaptic weights with our STDP rule in Eq ., ( 5 ) drive the initially random firing of the circuit in response to an input towards learning the correct generative model of the input distribution ., Whenever a neuron fires in response to , the STDP rule increases the weights of synapses from those presynaptic neurons that had fired shortly before ., In absence of a recent presynaptic spike from the weight is decreased ., As a consequence , when next a pattern similar to is presented , the probability for the same to fire and further adapt its weights , is increased ., Since becomes more of an “expert” for one subclass of input patterns , it actually becomes less likely to fire for non-matching patterns ., The competition in the WTA circuit ensures that other -neurons learn to specialize for these different input categories ., In the framework of Expectation Maximization , the generation of a spike in a -neuron creates a sample from the currently encoded posterior distribution of hidden variables , and can therefore be viewed as the stochastic Expectation , or -step ., The subsequent application of STDP to the synapses of this neuron can be understood as an approximation of the Maximization , or -step ., The online learning behavior of the network can be understood as a stochastic online EM algorithm ., We have previously shown that the output spikes of the WTA circuit represent samples from the posterior distribution in Eq ., ( 11 ) , which only depends on the ratios between the membrane potentials ., The rate at which these samples are produced is the overall firing rate of the WTA circuit and can be controlled by modifying the common inhibition of the neurons ., Although any time-varying output firing rate produces correct samples from the posterior distribution in Eq ., ( 11 ) of , for learning we also require that the input patterns observed at the spike times are unbiased samples from the true input distribution ., If this is violated , some patterns coincide with a higher , and thus have a stronger influence on the learned synaptic weights ., In Methods we formally show that acts as a multiplicative weighting of the current input , and so the generative model will learn a slightly distorted input distribution ., An unbiased set of samples can be obtained if is independent of the current input activation , e . g . if is constant ., This could in theory be achieved if we let depend on the current values of the membrane potentials , and set ., Such an immediate inhibition is commonly assumed in rate-based soft-WTA models , but it seems implausible to compute this in a spiking neuronal network , where only spikes can be observed , but not the presynaptic membrane potentials ., However , our results show that a perfectly constant firing rate is not a prerequisite for convergence to the right probabilistic model ., Indeed we can show that it is sufficient that and are stochastically independent , i . e . is not correlated to the appearance of any specific value of ., Still this might be difficult to achieve since the firing rate is functionally linked to the input by , but it clarifies the role of the inhibition as de-correlating from the input , at least in the long run ., One possible biologically plausible mechanism for such a decorrelation of and is an inhibitory feedback from a population of neurons that is itself excited by the neurons ., Such WTA competition through lateral inhibition has been studied extensively in the literature 16 , 33 ., In the implementation used for the experiments in this paper every spike from the -neurons causes an immediate very strong inhibition signal that lasts longer than the refractory period of the spiking neuron ., This strong inhibition decays exponentially and is overlaid by a noise signal with high variability that follows an Ornstein-Uhlenbeck process ( see “Inhibition Model in Computer Simulations” in Methods ) ., This will render the time of the next spike of the system almost independent of the value of ., It should also be mentioned that a slight correlation between and may be desirable , and might also be externally modulated ( for example through attention , or neuromodulators such as Acetylcholin ) , as an instrument of selective input learning ., This might lead e . g . to slightly higher firing rates for well-known inputs ( high ) , or salient inputs , as opposed to reduced rates for unknown arbitrary inputs ., In general , however , combining online learning with a sampling rate that is correlated to may lead to strange artifacts and might even prohibit the convergence of the system due to positive feedback effects ., A thorough analysis of such effects and of possible learning mechanisms that cope with positive feedback effects is the topic of future research ., Our theoretical analysis sheds new light on the requirements for inhibition in spiking WTA-like circuits to support learning and Bayesian computation ., Inhibition does not only cause competition between the excitatory neurons , but also regulates the overall firing rate of the WTA circuit ., Variability in does not influence the performance of the circuit , as long as there is no systematic dependence between the input and ., In our previous analysis we have assumed a simplified non-additive step-function model for the EPSP ., This allowed us to describe all input evidence within the last time window of length by one binary vector , but required us to assume that no two neurons within the same group fired within that period ., We will now give an intuitive explanation to show that this restriction can be dropped and present an interpretation for additive biologically plausibly shaped EPSPs as inference in a generative model ., The postsynaptic activation under an additive EPSPs is given by the convolution ( 17 ) where describes an arbitrarily shaped kernel , e . g . an -shaped EPSP function which is the difference of two exponential functions ( see 37 ) with different time constants ., We use 1 ms for the rise and 15 ms for the decay in our simulations ., replaces in Eq ., ( 2 ) in the computation of the membrane potential of our model neurons ., We can still understand the firing of neurons in the WTA circuit according to the relative firing probabilities in Eq ., ( 4 ) as Bayesian inference ., To see this , we imagine an extension of the generative probabilistic model in Fig . 1B , which contains multiple instances of , exactly one for every input spike from all input neurons ., For a fixed common hidden cause , all instances of are conditionally independent of each other , and have the same conditional distributions for each ( see Methods for the full derivation of the extended probabilistic model ) ., According to the definition in Eq ., ( 8 ) of the population code every input spike represents evidence that in an instance should take on a certain value ., Since every spike contributes only to one instance , any finite input spike pattern can be interpreted as valid evidence for multiple instances of inputs ., The inference of a single hidden cause in such extended graphical model from multiple instances of evidence is relatively straightforward: due to the conditional independence of different instances , we can compute the input likelihood for any hidden cause simply as the product of likelihoods for every single evidence ., Inference thus reduces to counting how often every possible evidence occurred in all instances , which means counting the number of spikes of every ., Since single likelihoods are implicitly encoded in the synaptic weights by the relationship , we can thus compute the complete input likelihood by adding up step-function like EPSPs with amplitudes corresponding to ., This yields correct results , even if one input neuron spikes multiple times ., In the above model , the timing of spikes does not play a role ., If we want to assign more weight to recent evidence , we can define a heuristic modification of the extended graphical model , in which contributions from spikes to the complete input log-likelihood are linearly interpolated in time , and multiple pieces of evidence simply accumulate ., This is exactly what is computed in in Eq ., ( 17 ) , where the shape of the kernel defines how the contribution of an input spike at time evolves over time ., Defining as the weight for the evidence of the assignment of to value , it is easy to see ( and shown in detail in Methods ) that the instantaneous output distribution represents the result of inference over causes , given the time-weighted evidences of all previous input spikes , where the weighting is done by the EPSP-function ., Note that this evidence weighting mechanism is not equivalent to the much more complex mechanism for inference in presence of uncertain evidence , which would require more elaborate architectures than our feed-forward WTA-circuit ., In our case , past evidence does not become uncertain , but just less important for the inference of the instantaneous hidden cause ., We can analogously generalize the spike-triggered learning rule in Eq ., ( 5 ) for continuous-valued input activations according to Eq ., ( 17 ) : ( 18 ) The update of every weight is triggered when neuron , i . e . the postsynaptic neuron , fires a spike ., The shape of the LTP part of the STDP curve is determined by the shape of the EPSP , defined by the kernel function ., The positive part of the update in Eq ., ( 18 ) is weighted by the value of at the time of firing the postsynaptic spike ., Negative updates are performed if is close to zero , which indicates that no presynaptic spikes were observed recently ., The complex version of the STDP curve ( blue dashed curve in Fig . 1B ) , which resembles more closely to the experimentally found STDP curves , results from the use of biologically plausible -shaped EPSPs ., In this case , the LTP window of the weight update decays with time , following the shape of the -function ., This form of synaptic plasticity was used in all our experiments ., If EPSPs accumulate due to high input stimulation frequencies , the resulting shape of the STDP curve becomes even more similar to previously observed experimental data , which is investigated in detail in the following section ., The question remains , how this extension of the model and the heuristics for time-dependent weighting of spike contributions affect the previously derived theoretical properties ., Although the convergence proof does not hold anymore under such general conditions we can expect ( and show in our Experiments ) that the network will still show the principal behavior of EM under fairly general assumptions on the input: we have to assume that the instantaneous spike rate of every input group is not dependent on the value of that it currently encodes , which means that the total input spike rate must not depend on the hidden cause ., Note that this assumption on every input group is identical to the desired output behavior of the WTA circuit according to the conditions on the inhibition as derived earlier ., This opens up the possibility of building networks of recursively or hierarchically connected WTA circuits ., Note also that the grouping of inputs into different is only a notational convenience ., The neurons in the WTA circuit do not have to know which inputs are from the same group , neither for inference nor for learning , and can thus treat all input neurons equally ., In biological STDP experiments that induce pairs of pre- and post-synaptic spikes at different time delays , it has been observed that the shape of the plasticity curve changes as a function of the repetition frequency for those spike pairs 40 ., The observed effect is that at very low frequencies no change or only LTD occurs , a “classical” STDP window with timing-dependent LTD and LTP is observed at intermediate frequencies around 20 Hz , and at high frequencies of 40 Hz or above only LTP is observed , independently of which spikes comes first ., Although our theoretical model does not explicitly include a stimulation-frequency dependent term like other STDP models ( e . g . 53 ) , we can study empirically the effect of a modification of the frequency of spike-pairing ., We simulate this for a single synapse , at which we force pre- and post-synaptic spikes with varying time differences , and at fixed stimulation frequencies of either 1 Hz , 20 Hz , or 40 Hz ., Modeling EPSPs as -kernels with time constants of 1 ms for the rise and 15 ms for the decay , we obtain the low-pass filtered signals as in Eq ., ( 17 ) , which grow as EPSPs start to overlap at higher stimulation frequencies ., At the time of a post-synaptic spike we compute the synaptic update according to the rule in Eq

Introduction, Results, Discussion, Methods

The principles by which networks of neurons compute , and how spike-timing dependent plasticity ( STDP ) of synaptic weights generates and maintains their computational function , are unknown ., Preceding work has shown that soft winner-take-all ( WTA ) circuits , where pyramidal neurons inhibit each other via interneurons , are a common motif of cortical microcircuits ., We show through theoretical analysis and computer simulations that Bayesian computation is induced in these network motifs through STDP in combination with activity-dependent changes in the excitability of neurons ., The fundamental components of this emergent Bayesian computation are priors that result from adaptation of neuronal excitability and implicit generative models for hidden causes that are created in the synaptic weights through STDP ., In fact , a surprising result is that STDP is able to approximate a powerful principle for fitting such implicit generative models to high-dimensional spike inputs: Expectation Maximization ., Our results suggest that the experimentally observed spontaneous activity and trial-to-trial variability of cortical neurons are essential features of their information processing capability , since their functional role is to represent probability distributions rather than static neural codes ., Furthermore it suggests networks of Bayesian computation modules as a new model for distributed information processing in the cortex .

How do neurons learn to extract information from their inputs , and perform meaningful computations ?, Neurons receive inputs as continuous streams of action potentials or “spikes” that arrive at thousands of synapses ., The strength of these synapses - the synaptic weight - undergoes constant modification ., It has been demonstrated in numerous experiments that this modification depends on the temporal order of spikes in the pre- and postsynaptic neuron , a rule known as STDP , but it has remained unclear , how this contributes to higher level functions in neural network architectures ., In this paper we show that STDP induces in a commonly found connectivity motif in the cortex - a winner-take-all ( WTA ) network - autonomous , self-organized learning of probabilistic models of the input ., The resulting function of the neural circuit is Bayesian computation on the input spike trains ., Such unsupervised learning has previously been studied extensively on an abstract , algorithmical level ., We show that STDP approximates one of the most powerful learning methods in machine learning , Expectation-Maximization ( EM ) ., In a series of computer simulations we demonstrate that this enables STDP in WTA circuits to solve complex learning tasks , reaching a performance level that surpasses previous uses of spiking neural networks .

circuit models, developmental neuroscience, synaptic plasticity, computational neuroscience, biology, computational biology, neuroscience

journal.ppat.1002403

Late Repression of NF-κB Activity by Invasive but Not Non-Invasive Meningococcal Isolates Is Required to Display Apoptosis of Epithelial Cells

The exclusive human bacterium Neisseria meningitidis ( the meningococcus ) is a major cause of infectious diseases worldwide , including meningitis and fulminant sepsis that are associated with significant morbidity and case fatality rates ranging from 10 to 50% in patients with severe septicaemia 1 , 2 and up to 20% of survivors sustain neurological sequelae 3 ., Despite this notoriety , N . meningitidis is a frequent inhabitant of the nasopharyngeal mucosa being asymptomatically carried by 10–35% of the adult population 4 , 5 ., A combination of bacterial factors and host susceptibility ( including age , prior viral infection , and genetic polymorphisms 6–8 ) , may ultimately lead to meningococcal disease ., Multilocus sequence typing ( MLST ) has been used to characterize the genotypes of meningococcal isolates determined by the sequence types ( STs ) and grouping these genotypes into distinct phylogenetic lineages referred to as “clonal complexes” 9 ., Comparisons of the genotypes of meningococcal isolates have shown that asymptomatic carriage isolates are genetically and antigenically highly diverse , whereas most meningococcal disease is caused by a limited number of clonal complexes known as the “hyper-invasive clonal complexes” 10–13 ., Genomic analysis failed to identify which bacterial features are responsible for the different epidemiologies 14 ., Moreover , bacterial virulence factors such as pili and capsule , although important in the establishment of the disease , are widely distributed among carriage and invasive isolates ., Therefore , a better understanding of the pathogenesis of this disease , notably the interaction with host cells , is central in developing new anti-meningococcal strategies ., There is increasing evidence that invasive meningococcal infections lead to cytopathic effects 15–20 ., These observations are consistent with the extensive cell injury and tissue damage seen in autopsy material from cases of human disease 21 ., We have recently shown a strong association of cytopathic effect to epithelial cells with isolates belonging to the hyper-invasive clonal complex ST-11 ., Infected cells presented features of apoptosis ., The apoptotic pathway induced by these isolates is mediated in part by lipooligosaccharides ( LOS ) , the major bacterial endotoxin , and involved tumor necrosis factor alpha ( TNF-α ) signaling through its cognate receptor TNFR1 ., In contrast , carriage isolates interfered with TNF-α-dependent apoptotic signaling by increasing extracellular shedding of TNFR1 leading to attenuation of the biological activity of TNF-α 22 ., Several signaling pathways are known to regulate apoptosis , but the transcription factor NF-κB lies at the nexus of both anti-apoptotic and proinflammatory cascades ( reviewed in references 23 , 24 ) ., In resting cells , NF-κB is sequestered in the cytosol through interactions with its inhibitor , IκB ., Proinflammatory stimuli , such as lipopolysaccharide ( LPS ) and TNF-α , activate a signaling pathway that results in phosphorylation and subsequent degradation of IκB by the proteasome machinery ., The liberated NF-κB translocates then to the nuclear compartment , where it activates the transcription of both proinflammatory and anti-apoptotic genes 25 ., Given the role of NF-κB in both inflammation and apoptosis , it is not surprising that certain pathogens have also evolved mechanisms to modulate NF-κB activity during infection 26 ., In this work we aimed to explore the differential ability of meningococcal invasive and non-invasive isolates to modulate the NF-κB activity ., We have shown that apoptosis of epithelial cells promoted by the ST-11 isolates is partially dependent on the meningococcal lipooligosaccharide LOS 22 ., Indeed , both invasive ST-11 isolates and the non-invasive carriage isolates were able to induce the expression of TLR4 at the surface of Hec-1B epithelial cells ( Supporting Figure S1 ) ., However , only invasive ST-11 isolates but not non-invasive carriage isolates or LOS-devoid mutants were able to induce apoptosis in Hec-1B epithelial cells ( Figure 1C and Table 1 ) ., Anti-TLR4 neutralizing mAb but not an isotype-matching IgG control Ab was able to inhibit apoptosis in epithelial cells infected by the ST-11 invasive isolates ( Table 1 ) ., Furthermore , the induction of apoptosis was abolished when TLR4 was knocked-down by siRNA-mediated TLR4 silencing strategy using specific TLR4 silencing duplex oligonucleotides siTLR4-1 or siTLR4-2 but not a non-specific control oligonucleotide ( siCTRL ) ( Supporting Table S1 and Figures 1A and 1B ) ., The induction of apoptosis by the invasive isolate LNP19995 was correlated with a significant high level of caspases-3 activity that was significantly reduced in the isogenic LOS-devoid mutant Z0305 or upon siRNA-mediated TLR4 silencing ( Figure 1D ) ., Caspase-3 activity in cells infected with the carriage isolate or its isogenic LOS mutant AD1001 were comparable to the background level ., Altogether , these data demonstrate that TLR4 is required to promote apoptosis by the ST-11 meningococcal isolates ., The following experiments were performed using LNP19995 and LNP21019 as a representative candidate of each , the invasive ST-11 isolates and the non-invasive carriage isolates respectively , unless otherwise specified ., TLR4 links to both MyD88 and TRIF to transduce signals to downstream effectors 27 ., MyD88 has an NH2-terminal death domain which links it to downstream effectors in the TLR signaling pathways and a COOH-terminal TIR domain which interacts with the cytoplasmic portion of various TLRs ., Each domain expressed alone functions as a dominant negative form 28 ., The TIR domain of TRIF has similar behaviour 29 , 30 ., To further explore the extent of LOS signaling pathway downstream TLR4 in the apoptosis promoted by the invasive ST-11 isolates , Hec-1B cells were knocked-down for functional MyD88 or TRIF by transfecting either TIR domain and apoptosis was analyzed after 9 h of infection in comparison with cells transfected with the empty vector control ., Expression of AU1-tagged dnMyD88 or Myc-tagged dnTRIF was confirmed in the transfected cells by immunoblotting using specific Abs directed against each tag ( Figure 2A ) ., As expected , the apoptotic level promoted by the wild-type ( WT ) ST-11 isolate LNP19995 in empty vector-transfected cells was dramatically decreased after infection with the LOS-deficient isogenic mutant Z0305 ., Comparably to TLR4 knock-down , expression of the dnMyD88 considerably impeded apoptosis promoted by the WT ST-11 isolate LNP19995 , while expression of the dnTRIF did not improve the survival of LNP19995-infected cells ( Figure 2B , 11 . 70±5 . 03% of dnMyD88-transfected cells underwent apoptosis while 37 . 01±2 . 28% of pcDNA3 or 31 . 58±2 . 98% of dnTRIF cells were already apoptotic ) ., Both transfected constructs ( dnMyD88 or dnTRIF ) resulted in apoptotic level in cells infected with the non invasive carriage isolate LNP21019 comparable to uninfected cells ( Figure 2B ) ., Results similar to dnMyD88 were obtained in cells transfected with a dnIRAK1 , an effector protein downstream MyD88 ( data not shown ) ., Taken together , our results indicate that LOS-mediated apoptotic signaling elicited by the ST-11 isolates through TLR4 involved a MyD88- but not TRIF-dependent pathway ., The subsequent steps in TLR4/MyD88 signalling lead to the activation of NF-κB which translocates into the nucleus to activate pro-inflammatory and pro-survival gene expression including TNF-α 31 ., We therefore sought to determine the role of NF-κB activity in apoptosis induced by the meningococcal ST-11 isolates ., We first followed the kinetic of NF-κBtrans-activation in response to meningococcal infection ., For that purpose , Hec-1B cells were transiently transfected with the plasmid ( Igκ ) 3conaLuc , where expression of the luciferase reporter gene is driven by an NF-κB-dependent promoter ., Luciferase activity normalized to the constitutive β-galactosidase activity control was assayed in a time course infection ., As depicted in Figure 3A , infection of Hec-1B cells with either isolates induced luciferase activity which peaked at 4h post-infection to nearly 20 fold relative to uninfected cells ., This early activation required LOS and occurred in TLR4-dependent manner as lack of LOS from both , the invasive or the carriage isolates ( Z0305 or AD1001 mutants , respectively ) or silencing of TLR4 considerably reduced transactivation of NF-κB ( Figure 3B ) ., Surprisingly , NF-κB-dependent luciferase activity decreased beyond 6 h of challenge with the invasive ST-11 isolates while persisted in response to infection with the carriage isolates ( Figure 3A ) ., These data corroborate with the EMSA assays ., Indeed , the DNA-binding activity of NF-κB to a specific radio-labeled probe peaked transiently at 4 h of infection then decreased beyond 6 h of infection with the ST-11 invasive isolate LNP19995 , while persisted longer in cells infected with the carriage isolate LNP21019 ( Supporting Figure S2 ) ., Interestingly the decrease of NF-κB-dependent luciferase and DNA-binding activities in cells infected with the ST-11 isolates was concomitant to induction of apoptosis by these isolates ., Collectively , these data show a differential modulation of NF-κB-dependent transcriptional activity between the invasive ST-11 and the non-invasive carriage isolates ., LOS purified from the invasive LNP19995 or the carriage LNP21019 isolates were able to trigger a persistent transactivation of NF-κB similarly to the carriage isolates , as judged by EMSA and luciferase reporter assays ., No alteration of NF-κB activity was observed at least up to 9 h of treatment with the purified LOS of both isolates ( Supporting Figure S3 ) ., In absence of serum , the lack of PorB expression in the mutant NM0401 resulted in slight decrease of the early activation of NF-κB comparing to the parental ST-11 strain LNP19995 ., In contrast to the PorB mutant , the LOS-devoid mutant Z0305 or the mutant lacking both LOS and PorB were strongly affected ( Figure 3C ) ., Interestingly , activation of NF-κB was decreased in the later time points in cells infected with the PorB mutant NM0401 similarly to cells infected with the wild type parental strain LNP19995 ( Figure 3C ) ., Overall , our data suggest that LOS is a potent activator of NF-κB comparing to PorB ., However , the late reduction of NF-κB activity provoked by the invasive ST-11 isolates seems to be independent of the expression of both PorB and LOS ., It is worth to note that expression of NF-κBp65 and p50 subunits was not affected over the time of infection with each isolates , excluding the modulation of NF-κB transcriptional activity due to the alteration of NF-κB expression ( Supporting Figure S4 ) ., To further explore the role of NF-κB in the apoptosis induced by the meningococcal ST-11 isolates , early activation of NF-κB was blocked 1 h prior to infection , using MG-132 inhibitor ., MG-132 is a peptide-aldehyde protease inhibitor that blocks NF-κB activation via inhibition of the proteasome function 32 , 33 ., Hec-1B cells were then infected for 9 h in presence of a non-toxic concentration of MG-132 or the carrier solvent DMSO ( 0 . 1% ) ., Neither MG-132 nor DMSO alone resulted in apoptosis above background levels in uninfected cells ( data not shown ) ., Unexpectedly , pre-treatment of cells with MG-132 effectively protected cells from apoptosis brought about infection with the invasive isolate comparing to cells infected in presence of DMSO ( Figure 4A ) ., Similar results were obtained using the specific NF-κB inhibitor IKK-Nemo Binding Domain ( NBD ) peptide ( data not shown ) ., Early activation of NF-κB seems then to be a pre-requisite to promote apoptosis of epithelial cells with the invasive isolates ., This requirement contrasts the cytoprotective role of NF-κB reported by several groups 34–38 ., Nevertheless , the kinetic of NF-κB transactivation reported in Figure 3A showed that both invasive and non-invasive isolates are able to promote NF-κB activity after 4h of infection ., However , only invasive ST-11 isolates are able to induce apoptosis after 9h of infection that was correlated with reduced NF-κB activity ., Non-invasive isolates promoted sustained NF-κB activity that correlated with protection against apoptosis ( Figure 3 ) ., We therefore explored the extent of the late down-regulation of NF-κB activity on the apoptotic cell death induced by the ST-11 invasive isolates ., Hec-1B cells were transfected with a FLAG tagged-constitutively active form of IKK2 ( CA-IKK2 ) , leading to a constitutive phosphorylation and degradation of IκBα and subsequent increase of NF-κB activity , 39 or pCMV2 empty vector control ( Figure 4B , immunoblot insert ) ., Expression of CA-IKK2 was able to abrogate LNP19995-induced apoptosis by four fold comparing to empty vector-transfected cells ( Figure 4B ) ., These results strongly suggest that the late down regulation of NF-κB activity is required for invasive ST-11 isolates to promote LOS-mediated apoptosis of epithelial cells ., One possible explanation for this dual effect of NF-κB activation is that early activation of NF-κB is required to promote expression of TNF-α , which acts lately to promote apoptosis of epithelial cells following the decrease of NF-κB activity ., Indeed , MG-132 dramatically reduced the level of secreted TNF-α with respect to DMSO-treated cells ( Supporting Figure S5 ) ., To test this hypothesis , MG-132 was incorporated after 4 h of bacterial challenge ( the period of which NF-κB transactivation was peaked ) ., In these conditions , apoptosis was strongly induced irrespective to infection with the invasive or the carriage isolates ( Figure 4C ) ., The presence of anti-TNF-α neutralizing antibody significantly abrogated this effect ( Figure 4C ) ., To further analyse the role of TNF-α in meningococcal induced apoptosis , TLR4-mediated NF-κB activation was blocked by transfecting cells with the dnMyD88-expressing vector ., Indeed , expression of the dnMyD88 as MG-132 pre-treatment strongly reduced the levels of secreted TNF-α in cells infected with both the invasive or the carriage isolates when compared to cells transfected with the empty vector control ( Supporting Figure S5 ) ., The apoptotic level promoted by the invasive isolate LNP19995 also decreased significantly in cells transfected with the dnMyD88 ( Figure 4A ) ., Addition of TNF-α rescued apoptosis regarding cells transfected with the dnMyD88 and infected in absence of TNF-α ., In all tested conditions , the carriage isolate induced low apoptosis similar to the background level ( Figure 4A ) ., These results underline the key role of TNF-α in promoting apoptosis by the invasive , but not the carriage isolates ., In contrast to LOS , TNF-α also activates NF-κB but in MyD88-independent manner 40 and Figure 4D ., After 9 h of incubation , TNF-α-induced NF-κB activity was also deceased upon infection with the invasive isolate LNP19995 but not with the carriage isolate LNP21019 ., Collectively , these data suggest that TNF-α secreted early upon infection with the ST-11 isolates is required to sensitize cells to apoptosis in synergy with the late down-regulation of NF-κB transcriptional activity ., We have previously shown that carriage non-invasive isolates inhibit TNF-α-dependent apoptotic pathway through increasing the shedding of soluble TNFR1 ( sTNFR1 ) which interfere with the biological activity of TNF-α ., Increased shedding of sTNFR1 occurred concomitantly to the decreased level of the membrane-associated TNFR1 ( mTNFR1 ) 22 ., Shedding of TNFR1 is mediated by TNF-α converting enzyme ( TACE also known as ADAM17 ) , a metalloproteinase localized to the cytoplasmic membrane 41 , 42 ., One possible explanation for the increased shedding of sTNFR1 in cells infected with the carriage isolates could be the increased activity of TACE/ADAM17 ., We therefore analyzed the ability of the specific TACE/ADAM17 inhibitor , TNF-α protease inhibitor–1 ( TAPI-1 ) , to block TNFR1 release in response to infection with the carriage isolates ., At the concentration employed ( 1 µM ) , TAPI-1 did not compromise cell viability as judged by PI staining ( data not shown ) ., As depicted in Figure 5 , cells infected for 9 h with the carriage isolates in presence of TAPI-1 failed to induce shedding of sTNFR1 comparing to DMSO-treated cells ( Figure 5 , left panel; 34 . 56±7 . 37 pg/ml for TAPI-treated cells vs . 433 . 3±58 . 75 pg/ml for DMSO-treated cells , P<0 . 001 ) and concomitantly displayed significant surface levels of membrane bound-TNFR1 ( Figure 5 , right panel; MFI 81 . 12±15 . 49 in TAPI-treated cells vs . 24 . 94±4 . 15 in DMSO-treated cells , P<0 . 001 ) ., TAPI-1 treatment further increased mTNFR1 level in cells infected with the ST-11 invasive isolates comparing to cells infected in presence of DMSO ( Figure 5 , right panel; MFI 146 . 4±35 . 79 vs . 73 . 88±5 . 48 , P<0 . 001 ) ., These data were also confirmed by immunofluorescence microscopy examination ( Supporting Figure S6 ) ., These results suggest that increased shedding of sTNFR1 in cells infected with the carriage non-invasive isolates involved TACE/ADAM17 activity ., Given that carriage and invasive ST-11 isolates differentially modulate the late activity of NF-κB , we sought to determine whether the modulation of NF-κB transcriptional activity may impact the surface display and extracellular shedding of TNFR1 ., To address this issue , MG-132 was added after 4 h of bacterial challenge to not compromise the early expression of TNF-α ., Comparing to DMSO vehicle-treated cells , MG-132-treatment led to significant decrease of sTNFR1 shedding and increase of mTNFR1 level in cells infected with the carriage isolates ( Figure 5 ) ., To test whether persistent transcriptional activity of NF-κB would inverse the surface display of TNFR1 due to infection with the ST-11 isolates , Hec-1B cells were transiently transfected with the plasmid pCA-IKK2 that was modified by insertion of EGFP marker to visualize transfected cells ., As control , cells were transfected with pcDNA3 harbouring the same marker ( empty vector ) ., As shown in Figure 6A , 39-47% and 35-40% of cells were transfected with empty vector control and pCA-IKK2 , respectively ( left panels ) ., After 9 h of infection , surface expression of TNFR1 ( mTNFR1 ) was examined in this sub-population of transfected cells ( GFP+ events gated in region R1 , Figure 6A , left panels ) ., As expected , among empty vector-transfected cells , infection with the ST-11 isolate LNP19995 resulted in higher level of mTNFR1 compared to cells infected with the carriage isolate LNP21019 ( MFI 67 . 13 vs . MFI 11 , respectively ) ., Interestingly , among pCA-IKK2-transfected cells , infection with the ST-11 isolate LNP19995 led to almost 50% decrease of TNFR1 surface level compared to empty vector-transfected cells ( Figure 6A , right panels ) ., Consistent with these results , the level of sTNFR1 significantly increased in CA-IKK2 transfected- , LNP19995-infected cells compared to empty vector-transfected , LNP19995 infected cells or cells infected with the carriage isolate LNP21019 ( Figure 6B ) ., Taken together , these results suggest that the differential modulation of NF-κB activity between the invasive ST-11 and the non invasive carriage isolates resulted in differential surface display of TNFR1 in TACE/ADAM17-dependent way ., MAP kinase ( MAPK ) signaling pathways are activated in inflammatory reactions and have been shown to play important role in cell growth and death 43 ., In general , p38 and c-Jun N-terminal kinase ( JNK ) are involved in cell death mechanisms , whereas Extracellular signal-regulated kinase ( Erk1/2 ) is critical for cell survival 44 ., We therefore investigated the potential role of MAP kinases in apoptosis of epithelial cells induced by the pathogenic meningococcal isolates ., First , the effect of meningococci on the activation of the above mentioned kinases was examined in time course infection using phospho-specific antibodies ., We monitored for total levels of each kinase to ensure that any change in phosphorylated protein levels was an actual measure of activation ., Phosphorylation of Erk1/2 increased after 2h of cell exposure to each meningococcal isolate and then decreased after 6 h ( Figure 7A , upper panel ) ., Phosphorylation of p38 MAP kinase increased beyond 6 h of infection and persisted thereafter ( Figure 7A , middle panel ) ., Interestingly , phosphorylation of JNK increased gradually up to 6 h post infection , and then decreased afterwards to almost reach the basal level in cells infected with the carriage non invasive isolate ., At the opposite , JNK phosphorylation was sustained in cells infected with the ST-11 isolate LNP19995 ( Figure 7A , lower panel ) ., As the carriage isolate , JNK was transiently activated upon treatment of Hec-1B cells with the LOS purified from the invasive or the carriage isolates ( Supporting Figure S3 ) ., These data corroborate with the sustained activation of NF-κB ., In all tested conditions , the total expression of each kinase was not altered ( Figure 7A ) ., To examine more thoroughly the role of the prolonged JNK activation in the cell death promoted by the invasive ST-11 isolates , we determined the effects of a dose range of the specific JNK inhibitor SP600125 45 on apoptotic cell death induced by the invasive isolate LNP19995 ., Pre-treatment of cells with SP600125 markedly subdued the extent of the invasive isolate LNP19995-mediated cell death in dose-dependent manner with a maximum effective dose of 140 nM ( Figure 7B , lower panel ) ., This dose had no significant effect on viability of uninfected cells ., In contrast to JNK , inhibition of p38 MAPK and Erk1/2 phosphorylation with their respective specific inhibitors had no beneficial effect on LNP19995-induced epithelial cell death and no effect on the viability of uninfected cells ( Figure 7B , middle and upper panels , respectively ) ., These data pointed out a selective differential ability of isolates to modulate the activation of JNK and suggest that infection with ST-11 invasive isolates induced death signals involving JNK , while attenuating survival signals in epithelial cells ., Previous observations reported that sustained JNK activation associated with inhibition of NF-κB activation , contributes to TNF-α-induced apoptosis 46 , 47 ., To determine the extent of the late reduction of NF-κB activity mediated by the ST-11 isolates on the sustained JNK activation , Hec-1B cells were transiently transfected with the CA-IKK2 and the activation of JNK was analyzed after 9 h of infection with the invasive isolate LNP19995 ., Comparing to the control empty vector-transfected cells , maintenance of NF-κB activation in CA-IKK2-transfected cells resulted in dramatic impairment of JNK phosphorylation ( Figure 8 ) that was associated with improvement of cell survival ( Figure 4C ) ., Our data establish a direct cause-effect of interference of ST-11 invasive isolates with NF-κB activity to allow a sustained JNK activation that is necessary to promote apoptosis of epithelial cells ., Isolates of several bacterial species such as N . meningitidis may exist as symbiote in their host but may also invade internal compartments of the host with important local and systemic inflammatory responses ., The consequences of the interaction with epithelial cells at the nasopharynx ( the portal of entry of N . meningitidis ) are therefore crucial in the pathophysiology of meningococcal infections ., Induction of cytokines and particularly TNF-α has been implicated in local disruption of epithelial barrier functions 48 , 49 and inducing epithelial cell apoptosis 50 ., This hallmark may pave the way for further meningococcal invasion and dissemination to deeper sites ( septicaemia and meningitis ) 19 ., Indeed , it has been reported that patients with meningococcal sepsis or meningitis often describe signs of pharyngitis before the onset of invasive disease while carriage isolates persist in the pharynx asymptomatically 51 ., We have previously reported the impressive ability of the invasive ST-11 meningococcal isolates to induce apoptosis of epithelial cells ., This overwhelming process , driven in part by the major bacterial endotoxin LOS , is dependent on an autocrin mechanism of TNF-α signaling through its receptor TNFR1 22 ., In contrast , infection with the non-invasive carriage isolates is associated with protection of epithelial cells mediated by the shedding of sTNFR1 resulting in alteration of the biological activity of TNF-α through soluble receptor-ligand complex formation and abrogation of apoptosis 22 ., In this work , we further identified the actors in the signalling pathways leading to this different behaviour of invasive and non invasive isolates ., Our data with siTLR4 approach confirm the role of TLR4 as a potential transducer of the LOS induction of apoptotic signalling as in other Gram-negative pathogens such as Yersinia and Salmonella 52 , 53 ., Our data further show that TLR4 signaling induced by N . meningitidis through occurs in MyD88-dependent manner 54 , 55 , but not through the TRIF adaptor molecule ( MyD88-independent pathway ) 54 , 56 ., Indeed , these results are in agreement with the previous reports showing that MyD88 and IRAK1 are involved in apoptotic signaling upon stimulation with bacteria or bacterial components 57 , 58 ., TRIF-dependent pathway ( MyD88-independent pathway ) is more involved in generating a type I IFN-dependent response that is essential to host defence against viral infection 54 , 56 ., Nevertheless , many TLRs signal through the adaptor protein MyD88 ., In this regard , our data cannot exclude the role of other TLRs ( which also signal through MyD88 such as TLR2 ) in the regulation of apoptosis upon infection with the pathogenic meningococcal isolates ., Expression of dn-MyD88 as TLR4 knock-down , considerably altered the apoptosis induced by the ST-11 isolates and dramatically reduced expression of TNF-α ., Moreover , our data suggest that the initial TLR4-dependent activation of NF-κB is required to establish apoptosis most likely through induction of TNF-αexpression ., Treatment with MG-132 after 4 h of infection promoted apoptosis in cells infected with the carriage isolates and further increased apoptosis induced by the ST-11 isolates ., However , sustained NF- κB activity seems to be protective against apoptosis as constitutive NF-κB activation mediated by the CA-IKK2 , rescued the viability of the ST-11-infected cells ., Interference with the NF-κB transcriptional activity to promote apoptosis of host cells has been described for other pathogens ., Uropathogenic E . coli ( UPEC ) was able to abrogate urothelial responses by blocking NF-κB translocation to the nucleus and by inhibiting NF-κB-dependent transcription in response to either LPS or TNF-α stimulation 59 ., Yersinia induces apoptotic cell death in macrophages by type III secretion system-mediated suppression of NF-κB activation 60 ., In our hands , expression level of NF-κB p65 and p50 subunits were not affected during meningococcal infection , excluding the possibility that invasive ST-11 isolates modulate the activity of NF-κB through alteration of its expression ., It has been shown that PorB activates NF-κB in TLR2-dependent manner 61 ., The late reduction of NF-κB activity seems to be independent on PorB expression although the early activation slightly decreased comparing to the wild type strain ., These results comfort our previous results showing that ST-11 isolates induce apoptosis in two independent pathways an extrinsic pathway promoted by LOS and an intrinsic pathway promoted by PorB 22 ., How ST-11 meningococcal isolates attenuate the activity of NF-κB to promote apoptosis of epithelial cells is currently under investigation ., Bacterial factor ( s ) may be responsible for this difference in the fate of NF- κB activity upon meningococcal infection ., As a biological consequence of the differential modulation of NF-κB activity between ST-11 and carriage isolates , cells exhibited:, i ) a differential display of TNFR1 expression at the surface of infected cells dependent on TACE/ADAM17 activity ( higher mTNFR1 and low sTNFR1 in cells infected with the ST-11 isolates versus low mTNFR1 and higher sTNFR1 in cells infected with the carriage isolates ) and, ii ) a differential profile of JNK activation ( sustained activation in cells infected with the ST-11 invasive isolates and transient activation in cells infected with the carriage non invasive isolates ) ., Indeed , we demonstrated that the inducible shedding of TNFR1 from cells infected with carriage isolates can be blocked by TAPI-1 , an inhibitor of TACE/ADAM17 ., Furthermore NF-κB inhibitor suppressed the shedding of TNFR1 in carriage isolates infected cells while constitutive activation of NF-κB resulted in increased TNFR1 shedding from ST-11-infected cells ., Our data are in agreement that sustained NF-κB activity is associated with up-regulation , maturation and increased activity of TACE/ADAM17/ADAM17 62 ., Nevertheless , the precise mechanism by which the meningococcal isolates modulate TACE/ADAM17 activity has yet to be identified ., Recently , the involvement of the MAPK pathway in response to infection by N . meningitidis has been reported ., It has been demonstrated that N . meningitidis can induce a sustained activation of p38 MAPK and JNK in vitro 63 ., However , the role of these MAPKs in meningococcal-induced cell death has not been documented ., It has been suggested that p38 and JNK are in general involved in cell death mechanisms , whereas Erk1/2 is critical for cell survival 64 ., Our findings provide evidence for the participation of sustained JNK phosphorylation in the regulation of epithelial cell apoptosis in response to infection by ST-11 meningococcal invasive isolates ., The JNK activity was only transiently detected with non invasive isolates ., Using microarray analysis , we have recently reported that TACE/ADAM17 expression is reduced in blood of infected mice with N . meningitidis ST-11 ( - 2 . 4 fold change and Z score = -1 . 7 ) 65 ., On the other hand , it has been shown that TACE/ADAM17 activity increases upon loss of c-Jun , a downstream target of JNK 66 ., Sustained JNK activation promoted by ST-11 isolates may therefore be involved in the increase of TNFR1 surface expression through activation of c-Jun ., Based on these results , our data lean toward a biphasic model to promote apoptosis by ST-11 isolates ( Figure 9 ) : First , LOS mediates early activation of NF-κB during meningococcal infection in TLR4/MyD88/IRAK1-dependent manner leading to induction of expression and early secretion of the pro-inflammatory cytokine TNF-α and its receptor TNFR1 ( Figure 9A ) ., The sustained NF- κB activity may then promote TACE/ADAM17 activity that allows shedding of TNFR1 and prevents a sustained activation of the apoptotic factor JNK ., This may then protect epithelial cells against apoptosis when they encounter non invasive carriage isolates ., At the opposite , invasive isolates may produce factor ( s ) that inhibit the NF- κB activity leading to the accumulation of membrane bound TNFR1 and a sustained activation of the apoptotic factor JNK ., ( Figure 9B ) ., N . meningitidis-host cell interaction seems to involve a complex process in which bacterial heterogeneity impact differentially on the modulation of host cell signaling ., Knowledge of the mechanism of alteration of NF-κB activity related to the detrimental effect of ST-11 invasive isolates may therefore provide better understanding and rational approaches for the control of invasive meningococcal infection ., RPMI 1640 , HBSS and trypsin-EDTA were obtained from Invitrogen ( France ) ., Cocktail of protease inhibitors and phenylmethylsulfonyl fluoride ( PMSF ) were from Boehringer Mannheim ( France ) ., Human TNF-α ( hTNF-α ) was

Introduction, Results, Discussion, Materials and Methods

Meningococcal invasive isolates of the ST-11 clonal complex are most frequently associated with disease and rarely found in carriers ., Unlike carriage isolates , invasive isolates induce apoptosis in epithelial cells through the TNF-α signaling pathway ., While invasive and non-invasive isolates are both able to trigger the TLR4/MyD88 pathway in lipooligosaccharide ( LOS ) -dependant manner , we show that only non-invasive isolates were able to induce sustained NF-κB activity in infected epithelial cells ., ST-11 invasive isolates initially triggered a strong NF-κB activity in infected epithelial cells that was abolished after 9h of infection and was associated with sustained activation of JNK , increased levels of membrane TNFR1 , and induction of apoptosis ., In contrast , infection with carriage isolates lead to prolonged activation of NF-κB that was associated with a transient activation of JNK increased TACE/ADAM17-mediated shedding of TNFR1 and protection against apoptosis ., Our data provide insights to understand the meningococcal duality between invasiveness and asymptomatic carriage .

Strains of Neisseria meningitidis isolated from patients induce apoptotic cell death through the TNF-α pathway , whereas strains isolated from healthy carriage isolates do not ., Part of the difference has been shown to arise from differential shedding of the type 1 TNF-α receptor ( TNFR1 ) from the surface of the cells infected with the carriage isolates ., Here , we elucidate some of the downstream signaling that differs between these isolates , specifically showing that carriage isolates induce sustained NF-κB activity , leading to cytoprotective events whereas invasive isolates block this NF-κB activity and thus fail to induce the downstream protective events .

medicine, infectious diseases, immunology, biology, microbiology, molecular cell biology

journal.pbio.1001968

The Chromatin Assembly Factor 1 Promotes Rad51-Dependent Template Switches at Replication Forks by Counteracting D-Loop Disassembly by the RecQ-Type Helicase Rqh1

The maintenance of genome stability requires a complex network to coordinate multiple pathways , including DNA replication , repair , and recombination in a chromatin context ., Replication stress , including obstacles to replication fork progression , has emerged as a major source of genome instability that fuels cancer development and underlies chromosome modifications observed in genomic disorders 1 ., Deciphering the control of repair pathways occurring at replication forks remains of crucial importance to understanding of the mechanisms underlying genome rearrangements ., Homologous recombination ( HR ) is an evolutionarily conserved mechanism that promotes DNA repair and contributes to accurate and complete DNA replication 2 ., When fork progression is disrupted by DNA damage or a fork obstacle , HR mediates the nascent strands to switch templates to resume DNA synthesis ., Template switch occurs either at the three-way branched junction of the fork to restart it or between sister-chromatids to fill in single-stranded DNA ( ssDNA ) gaps left behind the moving fork 3 , 4 ., This last pathway is referred to as error-free postreplication repair ( PRR ) 5 ., Faulty replication restart events are one of the causal mechanisms of genome instability ., When control of allelic recombination fails , nascent strands at a blocked replication fork can recombine with a nonallelic homologous repeat and initiate DNA synthesis on a noncontiguous template , thus resulting in the fusion of noncontiguous DNA segments and genome rearrangements 6–8 ., This mechanism is referred to as faulty template switch and is proposed to drive genome rearrangements in cancers cells ( e . g . , chromothripsis ) and in genomic disorders ( e . g . , complex rearrangements such as triplication-associated inversions ) 1 , 9 ., Faulty template switch between homologous repeats is reminiscent of nonallelic HR ( NAHR ) ., In yeast , inverted repeats are particularly prone to faulty template switching 10 ., Recently , both HR and error-free PRR have been reported as mechanisms of faulty replication leading to fusion of inverted repeats in human cells 11 ., Thus , the mechanisms of faulty template switch appear evolutionarily conserved ., HR has been extensively studied in the context of double-strand break ( DSB ) repair , but only a few studies have addressed the mechanisms of template switch 2 , 4 , 12 ., With the assistance of HR mediators ( such as Rad52 in yeast ) , the recombinase Rad51 nucleates onto ssDNA covered by RPA to form a nucleoprotein filament ., After the search for homology , the Rad51 filament invades a homologous DNA duplex to pair the invading ssDNA with the complementary strand , whereas the noncomplementary strand is displaced ., The resulted three-stranded intermediate is a type of joint molecule ( JM ) called a displacement loop ( D-loop ) in which the 3′ end of the invading strand primes DNA synthesis ., At replication forks , extension of the D-loop by DNA synthesis might permit the restoration of a functional replisome , thus ensuring the completion of DNA replication 13 ., In the context of DSB repair , the capture of the second DNA end results in the formation of a later JM called a double Holliday junction ( dHJ ) whose resolution by cleavage leads to crossover ( CO ) formation , a source of chromosome rearrangements associated with NAHR 1 , 12 , 14 ., Several DNA helicases/translocases have been shown to be involved in the prevention of mitotic CO by preventing D-loop formation or its disassembly—among them , Srs2 , FANCM , and RecQ-type helicases 15–23 ., Whether the outcomes of template switch at replication forks are also regulated by helicase-dependent D-loop dismantling is unknown ., HR occurs within DNA packaged into chromatin that needs to be disassembled and then restored after the recombination event is completed 24 ., Chromatin remodeling factors help in relaxing chromatin and in providing access to DNA damage signaling and repair machineries at damaged sites , but how chromatin restoration is coupled to HR remains poorly understood ., The chromatin assembly factor 1 , CAF-1 , is a histone H3-H4 chaperone that promotes DNA synthesis-coupled chromatin assembly during DNA repair and DNA replication 25–28 ., CAF-1 is a three-subunit complex conserved throughout evolution , and the three CAF-1 subunits in Schizosaccharomyces pombe are called Pcf1 ( SPBC29A10 . 03c ) , Pcf2 ( SPAC26H5 . 03 ) , and Pcf3 ( SPAC25H1 . 06 ) , which correspond , respectively , to the p150 , p60 , and p48 in mammalian cells 29 ., The large subunit of CAF-1 , p150 , interacts with PCNA , thus targeting CAF-1 to DNA synthesis sites at which CAF-1 and Asf1 ( anti-silencing factor 1 ) cooperatively assemble chromatin onto newly synthesized DNA in a PCNA-dependent manner 30–37 ., In response to DNA damage , the large subunit of CAF-1 and the heterochromatin factors HP1 ( heterochromatin protein 1 ) are targeted to mammalian HR foci , within which they promote the resection of DSBs and thus the recruitment of HR factors such as Rad51 38–41 ., After completion of DNA repair , CAF-1 and Asf1 restore nucleosomal organization at DNA damage 26–28 , 30 , 42–44 ., In budding yeast , CAF-1 and Asf1 are dispensable for DSB repair by HR but necessary for the restoration of the chromatin state , a step required to turn off checkpoint activation 45 , 46 ., It is suggested that CAF-1 primes DSB repair and then switches to an active histone chaperone mode to restore chromatin at DNA damage 24 ., Whether CAF-1 regulates other HR pathways such as template switch and whether it impacts repair fidelity is unknown ., Here , we identified that fission yeast CAF-1 acts in a HR pathway alternative to PRR when cells replicate a damaged template ., We revealed functional and physical interactions between CAF-1 and the RecQ-type helicase Rqh1 , the fission yeast BLM homologue ., Using a conditional replication fork obstacle , we report a novel chromatin factor-dependent step during HR-mediated template switch: CAF-1 counteracts the disassembly of D-loop intermediates by Rqh1 ., As a consequence , the likelihood of faulty template switch is controlled by the antagonistic roles of CAF-1 and Rqh1 in D-loop stability ., The protection of the D-loop requires the three CAF-1 subunits and its ability to interact with PCNA , showing that CAF-1 stabilizes the D-loop at the DNA synthesis step ., Thus , CAF-1 and Rqh1 act coordinately to maintain genome stability in response to replication stress ., We propose that CAF-1 plays a regulatory role during template switch by assembling chromatin on the D-loop and thereby impacting its resolution ., We wanted to ascertain whether CAF-1 was involved in replication-coupled DNA repair ., We focused on cell resistance to the alkylating agent methyl-methane sulfonate ( MMS ) that creates DNA lesions blocking fork elongation and known to induce template switch events 4 ., The deletion of pcf1 ( pcf1-d , SPBC29A10 . 03c ) did not affect cell sensitivity to MMS compared to wild-type ( wt ) cells ., However , combined with genetic backgrounds in which error-prone PRR ( the bypass of DNA lesions by translesion synthesis , rev1-d , SPBC1347 . 01c ) or error-free PRR ( rad8-d , SPAC13G6 . 01c , Rad8 being the homologue of budding yeast Rad5 ) were defective , pcf1-d resulted in an increased cell sensitivity to MMS , compared to each single mutant ( Figure 1 ) ., A similar genetic interaction was observed with Srs2 ( SPAC4H3 . 05 ) , a helicase involved in error-free PRR 47 ., These data suggest that CAF-1 acts in a replication-coupled DNA repair pathway but independently of the error-prone and error-free branches of PRR ., We then asked whether CAF-1 could act in the Rad51 ( SPAC644 . 14c ) -dependent HR pathway ., We found that the double mutant pcf1-d rad51-d exhibited only a modest increased sensitivity to MMS compared to the single mutant rad51-d ( Figure 1 ) ., These data indicate that CAF-1 may operate in the Rad51-dependent replication-coupled DNA repair pathway but may not function entirely through the HR pathway ., To decipher the role of CAF-1 in replication-coupled HR , we made use of a polar replication fork barrier ( RFB ) , which is genetically encoded by a DNA sequence called RTS1 bound by the protein Rtf1 whose expression is regulated at the transcriptional level via the use of the nmt41 promoter ., In the presence of thiamine in the media , Rtf1 is not expressed and the RTS1-RFB is not active; after at least 16 h following thiamine removal , Rtf1 is expressed and the RTS1-RFB is induced 48 ., In the t-ura4 <ori construct , a single RTS1-RFB is located near an efficient replication origin ( ori 3006/7 , on chromosome III ) to allow the block of forks emanating from this origin and moving toward the telomere , the main replication direction of the ura4 locus ( Figure S1A ) ., Blocked replication forks are restarted by HR and independently of DSBs 3 , 7 ., Perturbed replication-coupled chromatin assembly , due to CAF-1 and Asf1 deficiency , leads to a higher susceptibility of replication forks to collapse and thus an increased level of genome instability in budding yeast 49–52 ., We thus asked whether a defect in CAF-1 affects the activity of the RTS1-RFB and the early steps of HR at replication forks ., At the t-ura4 <ori locus , which contains a single fork barrier , the analysis of replication intermediates ( RIs ) by bidimensional gel electrophoresis ( 2DGE ) showed that the RTS1-RFB was as efficient in the absence of CAF-1 ( i . e . , in either pcf1-d , pcf2-d , or pcf3-d null mutant ) as in the wt strain ( Figure S1B–C ) ., Also , Rad52 , the main HR factor ( SPAC30D11 . 10 ) , was recruited to the fork barrier in the absence of CAF-1 to the same extent as in the wt strain ( Figure S1D ) ., Our data indicate that the RTS1-RFB was functional and prone to recruit HR factors in the absence of CAF-1 ., We then made use of another construct , t> ura4 <ori , that contains two RTS1 sequences integrated at both sides of ura4 ( Figure 2A ) 48 ., A third RTS1 sequence is present at its natural location , near the mat locus on the chromosome II ., Given the orientation of the RTS1 sequences relative to the main replication direction of each locus , the RTS1 sequence at the centromere ( cen ) -proximal side of ura4 behaves as a strong RFB , whereas the two other RTS1 sequences have poor RFB activity ., Occasionally during replication restart ( in ∼2%–3% of the cell population/generation ) , HR-mediated template switch results in nascent strands inappropriately invading the RTS1 sequence located in the vicinity of the blocked fork on chromosome III or the one located further away on chromosome II ., Such faulty template switches lead to chromosomal rearrangements including inversions and large palindromic chromosomes , as well as the loss of the ura4 marker ( Figures 2A and S2A ) 3 , 48 , 53 ., Importantly , chromosomal rearrangements were also observed in response to MMS treatment and in the absence of active RTS1-RFB , showing that conditional fork barriers are relevant models for the generation of rearrangements initiated by template switch 7 ., Also , replication restart and template switch-mediated rearrangements occur independently of PRR , making the RTS1-RFB assay particularly useful to decipher the role of CAF-1 in replication-coupled HR 7 , 53 ., In the t> ura4 <ori strain , fork arrest at the cen-proximal side of ura4 leads to an 11-fold increase in the loss of the ura4 marker ( Table 1 and 53 ) ., The loss of ura4 corresponds to the deletion of ura4 on chromosome III ( genomic deletion ) or translocation to chromosome II; each event can be distinguished by PCR ( Figure 2A–B ) ., Genomic deletion and translocation result from HR-mediated faulty template switches between the three dispersed RTS1 sequences ( Figure 2A ) 53 ., Consistent with this , both fork-arrest–induced genomic deletion and translocation are dependent on Rad52 ( Table 1 , Figure 2B , and 53 ) ., The rad52-d mutant experiences a loss of viability upon induction of the RTS1-RFB , a phenotype not observed in strains deficient for individual CAF-1 subunits ( Figure 2C ) ., However , we observed that a defect in CAF-1 leads to a 3- to 5-fold reduction in the rate of fork-arrest–induced ura4 loss , compared to the wt strain ( Table 1 ) ., PCR analysis showed that both genomic deletion and translocation induced by the active RTS1-RFB were affected: In the pcf1-d mutant , these events were reduced by 6- and 10-fold , respectively , compared to the wt strain ( p<0 . 0001 ) ( Figure 2B–D ) ., Our data indicate that , surprisingly , faulty template switch at blocked forks requires CAF-1 ., As Rad52 is efficiently recruited to the RTS1-RFB in the absence of CAF-1 ( Figure S1D ) , this suggests that CAF-1 promotes template switch downstream of the recruitment of HR factors ., In the t> ura4 <ori strain , fork arrest at the cen-proximal side of ura4 induces stalled nascent strands to switch template and to recombine with the opposite RTS1 sequence located on the telomere-proximal side of ura4 ( Figure 3A ) 3 ., This template switch event leads to a stable and early JM ( JM-A ) , which consists of a D-loop structure ., Then , the approaching opposite fork is stalled by the RTS1-RFB , leading to a second template exchange and the formation of a later JM ( JM-B ) , which contains HJ-like structures ( Figure 3A–B ) ., Thus , the D-loop structure is the precursor of the HJ-like intermediate ., Both types of JMs are detectable by 2DGE and are dependent on Rad52 3 ., The resolution of HJ-like structures leads to chromosomal rearrangements: acentric and dicentric isochromosomes or chromosomes in which ura4 has switched orientation ( Figures 3A and S2A ) ., Chromosomal rearrangements can be detected by pulse field gel electrophoresis ( PFGE ) and restriction fragment length analysis ( RFLA ) , followed by Southern blotting 3 ., We further investigated the role of CAF-1 during template switch by these physical assays ., The intensity of both types of JMs was severely decreased in strains deficient for individual CAF-1 subunits ( Figure 3B–C ) ., In addition , all types of chromosomal rearrangements , the products of resolution of HJ-like structures , were reduced by 2- to 3-fold in CAF-1 defective strains ( p<0 . 003 ) ( Figures 3D–E and S2B–C ) ., Thus , the decreased intensity of HJ-like structures could not be explained by a faster cleavage of these structures ., Because CAF-1 does not prevent HR factor recruitment at blocked forks , we rather envisioned that D-loop intermediates are formed but dismantled more quickly in the absence of CAF-1 , thus resulting in a decreased level of HJ-like intermediates ., To test this hypothesis , we analyzed genetic interactions with mus81 ( SPCC4G3 . 05c ) ., Fission yeast Mus81 is an endonuclease involved in the cleavage of HJs 54 ., As previously reported , HJ-like structures , but not D-loop intermediates , accumulated in mus81-d cells , thus resulting in cell death upon induction of the RTS1-RFB ( Figure S3 ) 3 ., Consistent with a faster dismantling of the D-loop and less HJ-like structures being produced in the absence of CAF-1 , the deletion of pcf1 rescued the sensitivity of mus81-d cells to the induction of the RTS1-RFB ( Figure S3A–B ) ., Analysis of JMs by 2DGE confirmed that the intensity of both JMs remained decreased in the double mutant compared to the single mutant mus81-d ( Figure S3C–D ) ., The data are consistent with the hypothesis that HJ-like structures are formed less often in the absence of CAF-1 due to faster dismantling of D-loop intermediates ., To reinforce this last hypothesis , we investigated genetic interaction with the recombinase rad51 required to promote D-loop formation ., In the absence of Rad51 , chromosome rearrangements are produced without D-loop formation , probably via the single strand annealing function of Rad52 3 ., We reasoned that if CAF-1 stabilizes Rad51-dependent D-loop intermediates , its function in promoting template switch should rely on a functional Rad51 pathway ., The type and level of chromosome rearrangements observed in the double mutant pcf1-d rad51-d was similar to those of the single rad51-d mutant , showing that rad51 and pcf1 are epistatic ( Figures 4A and S4A–B ) ., The data are consistent with CAF-1 acting in the Rad51 pathway to promote HR at replication forks , likely downstream of the formation of D-loop intermediates ., To extend our conclusion of CAF-1 acting in the Rad51 pathway to prevent D-loop disassembly , we performed genetic analysis ., In both fission and budding yeast models , the concomitant inactivation of a RecQ-helicase and Srs2 results in a pronounced slow growth phenotype or cell death , a phenotype rescued by the deletion of rad51 47 , 55 , 56 ., It has been proposed that this synthetic sickness/lethality results from the accumulation of unresolved JMs that impinge on cell fitness ., Deleting either pcf1 or pcf2 led to a marked rescue of the slow growth phenotype of the rqh1-d srs2-d strain , although to a less extent than the deletion of rad51 ( Figure 4B–C ) ., These data are consistent with CAF-1 acting in the Rad51 pathway and promoting template switch at replication forks by stabilizing D-loop intermediates ., We investigated the mechanism by which CAF-1 prevents D-loop dismantling ., Several helicases have been implicated in D-loop dissociation including Srs2 , Fml1 , and Rqh1 15–23 ., We found no evidence of Fml1 ( SPAC9 . 05 ) promoting template switch at the site-specific arrested fork ( unpublished data ) ., Given the synergistic sensitivity of the double mutant pcf1-d srs2-d to MMS , we first analyzed the interactions between CAF-1 and Srs2 using our model system for template switch ., We previously reported that Srs2 promotes JM formation and chromosomal rearrangements formed by template switch 3 ., We found that strains defective for both CAF-1 and Srs2 showed a reduced level of chromosome rearrangements , similar to those observed in each single mutant ( Figure S4 ) ., Thus , CAF-1 and Srs2 might act in the same pathway promoting template switch at replication forks ., The human RecQ helicase BLM and the large subunit of CAF-1 ( p150 ) physically interact to coordinately promote cell survival to replication stress 57 ., We found that the double mutant pcf1-d rqh1-d was more sensitive to MMS than the single rqh1-d mutant , rqh1 ( SPAC2G11 . 12 ) being the fission yeast homologue of BLM ( Figure 1 ) ., Also , pcf1-d rqh1-d was more sensitive to camptothecin ( CPT , a topoisomerase 1 inhibitor ) than each single mutant , whereas the deletion of pcf1 suppressed the sensitivity of the single mutant rqh1-d to hydroxyurea ( HU ) , a ribonucleotide reductase inhibitor that depletes dNTP pools and stalls replication forks ( Figure S5 ) ., Co-immunoprecipitation experiments showed that Rqh1 and Pcf1 physically interact ( Figures 5A and S6A ) ., Thus , functional interactions between CAF-1 and Rqh1 to promote cell resistance to replication stress are evolutionarily conserved ., RecQ helicases prevent genome instability by promoting the dissolution of early ( D-loop ) and late ( double HJs ) JMs 54 ., We previously have proposed that Rqh1 limits genome instability at replication forks by disassembling Rad51-dependent D-loops 3 ., In the RTS1-RFB assay , HJs formed between RTS1 repeats cannot branch migrate in vitro and thus cannot be resolved by dissolution ., Accordingly , HJ-like intermediates did not accumulate in rqh1-d cells compared to wt cells ( Figure 5B–C , panels 7 and 10 ) 3 ., We analyzed whether Rqh1 could be responsible for D-loop dismantling in the absence of CAF-1 ., In a pcf1-d rqh1-d and in a pcf2-d rqh1-d strain , the level of both JMs was restored to those observed in either rqh1-d or wt cells ( Figure 5B–C ) ., To verify that the stability of JMs are restored in vivo and does not result from an in vitro artifact during DNA manipulation , DNA samples were cross-linked prior to extraction ., In such conditions , the lack of JMs in the absence of CAF-1 was confirmed ( Figure 5B , panel 5 ) , showing that JMs are unstable in vivo ., Also , the intensity of JMs was restored to a wt level by deleting rqh1 ( Figure 5B , panels 6 and 9 ) , showing that Rqh1 is responsible for the lack of JMs in the absence of CAF-1 in vivo ., Consistently , both ura4 inversion and acentric chromosomes , the resolution products of HJ-like structures , were restored to wt levels in strains defective for CAF-1 and Rqh1 ( Figures 5D–E and S6B ) ., Our data establish that Rqh1 disassembles the D-loop in the absence of CAF-1 , and we propose that CAF-1 promotes template switch at the replication fork by counteracting D-loop disassembly by Rqh1 ., We analyzed the level of genomic deletion and translocation that result from faulty template switch between RTS1 sequences on chromosomes II and III 53 ., Following induction of the RTS1-RFB , deleting pcf1 resulted in a 6 and 10 times reduction in the rate of genomic deletion and translocation , respectively , compared to the wt strain ( Figure S6C–D ) ., In contrast , the rate of these events was reduced by only ∼1 . 8 times by deleting pcf1 in the absence of Rqh1 ( Table 1 and compare rqh1-d and pcf1-d rqh1-d strains on Figure S6C ) ., PCR analysis showed that translocation and deletion events were decreased in pcf1-d cells compared to wt cells ( Figure 2B ) , but not when rqh1 is deleted ( Figure S6D ) ., Altogether our data reveal that the likelihood of faulty template switch is controlled by the antagonistic roles of CAF-1 and Rqh1 in processing the D-loop ., CAF-1 mediates replication-coupled chromatin assembly and interacts with the heterochromatin factor Swi6 ( SPAC664 . 01c , the human HP1 homologue ) to assist the maintenance of heterochromatin and silencing during S-phase 29 ., A strain mutated for swi6 exhibited no defect in the accumulation of the acentric chromosome following the activation of the RTS1-RFB , indicating that the role of CAF-1 in template switch is unlikely to involve heterochromatin ( Figure S7 ) ., Deposition of histone H3-H4 onto newly synthesized DNA by CAF-1 requires , in vitro , its three subunits and its ability to interact with the replication factor PCNA ( SPBC16D10 . 09 ) 28 , 31–33 , 35 , 37 , 44 ., The three strains pcf1-d , pcf2-d , and pcf3-d exhibited a similar phenotype: fewer faulty template switches and a faster dismantling of the D-loop ( Figures 2 and 3 ) ., A strain in which the three subunits of CAF-1 have been inactivated showed a decreased level of acentric chromosomes , one of the products of JM resolution , similar to those observed in each single mutant ( Figure 3D–E ) ., Thus , the role of CAF-1 in promoting template switch is not specific to a single subunit but necessitates the three subunits to act in the same HR pathway ., In budding and fission yeast , the large subunit of CAF-1 contains only one canonical PCNA interacting peptide ( PIP box ) ., We mutated the key residues to alanine to generate a mutant of pcf1 unable to interact with PCNA ( pcf1-PIPmut ) ( Figure 6A ) ., Co-immunoprecipitation showed that mutating the PIP box of Pcf1 severely impaired the interaction of Pcf1 with PCNA without affecting its interaction with Pcf2 ( Figure S8A ) ., The interaction of Pcf2 with PCNA was also dependent on the PIP box of Pcf1 ( Figure S8A ) ., Thus , expressing Pcf1-PIPmut leads to the formation of a CAF-1 complex unable to interact with PCNA ., As expected , mutating the PIP box of Pcf1 led to a loss of Pcf1 foci in S-phase cells and a loss of co-localization with replication factories ( labeled with a CFP-tagged version of PCNA ) ( Figure S8B ) ., Thus , the canonical PIP box of Pcf1 is sufficient to target CAF-1 into replication foci and expressing Pcf1-PIPmut is likely to impair replication-coupled chromatin assembly by CAF-1 ., Then , we investigated the phenotype of the pcf1-PIPmut strain ., First , the stability of JMs was impaired and consistently the level of the acentric chromosome ( one of the products of JM resolution ) was reduced as in the pcf1-d strain ( Figure 6B , C , D ) ., Second , the rates of genomic deletion and translocation induced by the active RTS1-RFB were similarly decreased in pcf1-PIPmut and pcf1-d cells , compared to the wt strain ( Table 1 , Figure 6E–F ) ., Thus , mutating the PIP box of Pcf1 is sufficient to mimic the deletion of Pcf1 ., Thus , the role of CAF-1 in promoting template switch by preventing Rqh1-dependent dismantling of the D-loop requires the full complex and the capacity to interact with PCNA ., Beyond its role in chromatin restoration at DNA damage sites , roles for CAF-1 in recombinational DNA repair pathways have been reported 24 , 45 , 46 ., Budding yeast CAF-1 protects against DSBs by acting both in HR and nonhomologous end-joining pathways 58 , 59 ., A defect in CAF-1 also leads to a decreased efficiency of DSB-induced recombinational repair in drosophila 60 ., More recently , a genetic screen has identified CAF-1 as promoting break-induced replication , a one-ended invasion HR pathway that occurs when the homology between the broken end and the donor DNA molecules is limited to one broken arm 61 ., In mammals , CAF-1 acts in both the early and late steps of HR-mediated DNA repair by promoting the resection of DSBs and the recruitment of HR factors and then the restoration of chromatin state when repair is completed 24 ., Here , we report that CAF-1 promotes replication-coupled DNA repair independently of the error-prone and error-free branches of PRR ., The sealing of ssDNA gaps left behind moving forks involves template switches mediated by the error-free branch of PRR 4 , 5 ., This damage tolerance pathway requires Rad5 , Rad51 , and the ubiquitination of PCNA ., Our data place CAF-1 in an alternative Rad51-dependent template switch pathway ., Consistent with this , replication restart and chromosome rearrangements mediated by template switch at site-specific arrested forks occur independently of the ubiquitination of PCNA 7 , 53 ., We propose that CAF-1 acts in Rad51-dependent template switches occurring during replication restart ., We identified the underlying mechanism: CAF-1 stabilizes the D-loop by preventing its disassembly by the helicase Rqh1 ., Consequently , the likelihood of faulty template switch , a type of NAHR causing chromosomal rearrangements , is controlled by the antagonistic activities of CAF-1 and Rqh1 at the D-loop: CAF-1 stabilizing the D-loop and Rqh1 promoting its disassembly ., Functional interplays between CAF-1 and Rqh1 in response to replication stress are evolutionarily conserved ( see below ) ., In mammals , CAF-1 primes HR events at DNA damage by promoting the end-resection of DSBs and thus the recruitment of HR factors such as Rad51 24 ., Then , CAF-1 might switch towards its histone chaperone mode to restore chromatin after the completion of the HR event ., Here , we report a novel step at which CAF-1 promotes HR: By preventing D-loop disassembly , CAF-1 impacts the resolution of the subsequent HR event ., Thus , the role of CAF-1 during HR might be more dynamic than previously anticipated , not only acting in the early and final steps , but having potential roles all along the HR process ., We propose that CAF-1 , and potentially chromatin assembly coupled to the DNA synthesis step of the HR event , is an important regulatory point of template switch ., Defects in the RecQ-type helicase BLM lead to Blooms syndrome , a human disorder associating genomic instability and cancer predisposition ., Functional interactions between BLM and the p150 large subunit of CAF-1 have been previously reported in response to replication stress 57 ., Here , we identified that interplays between CAF-1 and BLM are evolutionarily conserved in fission yeast ., The large subunit of CAF-1 , Pcf1 , and Rqh1 physically interact and act in a coordinated way to promote survival and maintain genome stability in response to replication stress ., Importantly , we uncovered the underlying mechanism ., Using genetic and physical assays that allow the analysis of the individual steps of HR-mediated template switch at a single replication fork , we found that the impaired stability of D-loop intermediates due to a CAF-1 defect results from the activity of Rqh1 ., CAF-1 thus counteracts D-loop dismantling by Rqh1 ., The RecQ helicase family is also involved in the rescue and stability of stalled forks , however we excluded interplays between CAF-1 and Rqh1 in this process 62–64 ., First , the site-specific arrested fork is stable and prone to recombination events in both single and double mutants ., Second , CAF-1 acts downstream of D-loop formation by Rad51 ., We hypothesize that nucleosome assembly on the D-loop is promoted by the interaction of CAF-1 with PCNA and that the nucleosomal nature of the D-loop prevents disassembly by Rqh1 ., We cannot exclude that the interaction with PCNA simply serves to recruit CAF-1 to the D-loop where CAF-1 could either directly counteract Rqh1 action or trigger the recruitment of an additional factor counteracting Rqh1 activity ( Figure 6G ) ., In human cells , BLM inhibits CAF-1–mediated chromatin assembly coupled to DNA repair 57 ., Through physical interactions , Rqh1 could also mediate CAF-1 recruitment to the D-loop on which Rqh1 could inhibit chromatin assembly by CAF-1 ., However , such hypotheses are not sufficient to account for all our observations ., Indeed , in the absence of CAF-1 and of any potential histone deposition on JMs , the D-loop is disassembled faster by Rqh1 , thus rather suggesting a model in which CAF-1 counteracts Rqh1 activity ., Preventing Rqh1-dependent D-loop dismantling requires the three subunits of CAF-1 and its interaction with PCNA as for optimal histone deposition onto newly replicated DNA in vitro 25 , 31 , 37 ., Therefore , we propose that CAF-1 prevents D-loop disassembly by promoting histone deposition onto the D-loop ., We could not confirm this hypothesis by generating CAF-1 mutated forms unable to interact with histones , as CAF-1 binds histone H3-H4 by multiple interactions: Each subunit interacts directly with histones and independently of the two other subunits ., The human p150 interacts with histone H3-H4 via an acidic domain of 350 residues containing the KER and ED domains ., The third subunit p48 interacts with the N-terminal domain of histone H4 , and deleting this domain is not sufficient to abolish in vitro chromatin assembly 65 ., Thus , the complexity of the protein interface between CAF-1 and histones currently limits our ability to genetically impair the interaction of CAF-1 with histones ., Although the sole absence of CAF-1 does not confer cell sensitivity to MMS , our data place CAF-1 in a Rad51-dependent template switch pathway by stabilizing D-loop intermediates ., Thus , redundant pathways must exist for D-loop stabilization ., On the other hand , CAF-1 is critical for faulty template switch events occurring between repeated sequences , a type of NAHR ., Protection of the D-loop by CAF-1 during extension by DNA synthesis might provide a mechanism that allows the stabilization of the heteroduplex ., This CAF-1–dependent D-loop stabilization might be critical when the homology between DNA molecules is limited ( e . g . , in NAHR ) , but alternative mechanisms of stabilizing the heteroduplex likely exist in the case of allelic HR ., For example , the ability of Rad51 to branch migrate a single HJ behind the initial point of strand invasion provides the opportunity to extend the heteroduplex without DNA synthesis 66 ., Such a mechanism can operate when the two recombinant molecules share a substantial length of homology ., The confined length of homology in case of faulty template switch ( ∼900 bp for the RTS1 sequence compared to unconfined length of homology between sister chromatids ) might restrict the effectiveness of this process .,

Introduction, Results, Discussion, Materials and Methods

At blocked replication forks , homologous recombination mediates the nascent strands to switch template in order to ensure replication restart , but faulty template switches underlie genome rearrangements in cancer cells and genomic disorders ., Recombination occurs within DNA packaged into chromatin that must first be relaxed and then restored when recombination is completed ., The chromatin assembly factor 1 , CAF-1 , is a histone H3-H4 chaperone involved in DNA synthesis-coupled chromatin assembly during DNA replication and DNA repair ., We reveal a novel chromatin factor-dependent step during replication-coupled DNA repair: Fission yeast CAF-1 promotes Rad51-dependent template switches at replication forks , independently of the postreplication repair pathway ., We used a physical assay that allows the analysis of the individual steps of template switch , from the recruitment of recombination factors to the formation of joint molecules , combined with a quantitative measure of the resulting rearrangements ., We reveal functional and physical interplays between CAF-1 and the RecQ-helicase Rqh1 , the BLM homologue , mutations in which cause Blooms syndrome , a human disease associating genome instability with cancer predisposition ., We establish that CAF-1 promotes template switch by counteracting D-loop disassembly by Rqh1 ., Consequently , the likelihood of faulty template switches is controlled by antagonistic activities of CAF-1 and Rqh1 in the stability of the D-loop ., D-loop stabilization requires the ability of CAF-1 to interact with PCNA and is thus linked to the DNA synthesis step ., We propose that CAF-1 plays a regulatory role during template switch by assembling chromatin on the D-loop and thereby impacting the resolution of the D-loop .

Obstacles to the progression of DNA replication forks can result in genome rearrangements that are often observed in cancer cells and genomic disorders ., Homologous recombination is a mechanism of restarting stalled replication fork that involves synthesis of the new DNA strands switching templates to a second ( allelic ) copy of the DNA sequence ., However , the new strands can also occasionally recombine with nonallelic repeats ( distinct regions of the genome that resemble the correct one ) and thereby cause the inappropriate fusion of normally distant DNA segments; this is known as faulty template switching ., The chromatin assembly factor 1 ( CAF-1 ) is already known to be involved in depositing nucleosomes on DNA during DNA replication and repair ., We have found that CAF-1 is also involved in the recombination-mediated template switch pathway in response to replication stress ., Using both genetic and physical assays that allow the different steps of template switch to be analyzed , we reveal that CAF-1 protects recombination intermediates from disassembly by the RecQ-type helicase Rqh1 , the homologue of BLM ( people with mutations that affect BLM have Blooms syndrome , an inherited predisposition to genome instability and cancer ) ., Consequently , the likelihood of faulty template switch is controlled by the antagonistic activities of CAF-1 and Rqh1 ., We thus identified an evolutionarily conserved interplay between CAF-1 and RecQ-type helicases that helps to maintain genome stability in the face of replication stress .

dna electrophoresis, dna damage, mutation, fungi, dna replication, dna, epigenetics, molecular genetics, chromatin, dna synthesis, schizosaccharomyces, homologous recombination, chromosome biology, schizosaccharomyces pombe, nucleosomes, yeast, biochemistry, cell biology, genetics, biology and life sciences, dna repair, dna recombination, molecular cell biology, genetics of disease, organisms

A molecular switch for times of replication stress - Chromatin Assembly Factor 1 helps to protect DNA during recombination-mediated template-switching, favoring the rescue of stalled replication forks by both beneficial and detrimental homologous recombination events.

journal.ppat.1005785

Exaptation of Bornavirus-Like Nucleoprotein Elements in Afrotherians

Many eukaryotic genomes contain endogenous viral elements ( EVEs ) , which are derived from viruses 1 , 2 ., Retroviruses are known as a major source of EVEs , such that approximately 8% of the human genome consists of endogenous retroviruses ( ERVs ) 3 ., These EVEs not only serve as molecular fossil records representing the development of ancient to modern relationships between retroviruses and hosts , but also occasionally contribute to the evolution of hosts through exaptation 4–7 ., For example , the Syncytin genes derived from envelope genes of retroviruses are involved in placentation in mammals 6 ., In addition , some ERVs are involved in EVE-derived immunity ( EDI ) , which acts as an anti-virus factor against exogenous retrovirus infections 4 , 5 , 7 ., Recently , EVEs derived from non-retroviral DNA and RNA viruses were discovered in many eukaryotic genomes 8–12 ., An endogenous bornavirus-like element ( EBL ) was the first EVE identified to be derived from a non-retroviral RNA virus in mammalian genomes 10 ., EBLs are most closely related to bornavirus , which is a mononegavirus encoding a nucleoprotein ( N ) , phosphoprotein ( P ) , matrix protein ( M ) , glycoprotein ( G ) , RNA-dependent RNA polymerase ( L ) , and accessory protein ( X ) 13 , 14 ., To date EBLs derived from N , M , G , and L genes have been identified and designated as EBLN , EBLM , EBLG , and EBLL , respectively 9 ., Some EBL sequences are apparently initiated with a bornavirus transcription start site and ended with a polyA and flanked by target-site duplications ( TSDs ) , suggesting that they were generated through the LINE-1 machinery 8–12 ., EBLNs are found in the genomes of diverse animals including snakes , turtles , moles , rodents , primates , and afrotherians 8–17 ., ORFs found in EBLNs are in most cases fragmented by premature termination codons , although occasionally EBLNs with relatively long ORFs have been identified ., Haplorhini primates have maintained an EBLN encoding 366 amino acids ( aa ) , which is similar in length to bornavirus N ( 370 aa ) , for more than 40 million years ( MY ) ., However , no natural selection has been detected on primate EBLNs at the amino acid sequence level 15 ., Thirteen-lined ground squirrel ( Ictidomys tridecemlineatus ) has harbored an EBLN ( itEBLN ) encoding 207 aa for ~0 . 3 MY 18 ., A recombinant itEBLN protein conferred resistance to bornavirus infection on human cells by being incorporated into bornavirus particles as incompetent nucleoproteins 19 , suggesting a possibility that itEBLN might act as an EDI protein , which should be tested in thirteen-lined ground squirrel ., Overall , there is little evidence indicating that EBLNs encode a functional protein in their hosts ., Here , in a search for EBLNs that have been co-opted to encode functional proteins in the host , we investigated afrotherian EBLNs containing relatively long ORFs ., Afrotherians are a diverse group of mammals that originated in Africa 83 . 3 MY ago ( MYA ) 20 ., The superorder Afrotheria includes orders Proboscidea ( e . g . , elephant ) , Sirenia ( e . g . , dugong and manatee ) , Hyracoidea ( e . g . , hyracoid ) , Macroscelidea ( e . g . , elephant-shrew ) , Tubulidentata ( e . g . , aardvark ) , and Tenrecoidea ( e . g . , tenrec and golden mole ) 21 ., We present evidence for the existence of EBLNs encoding functional proteins and their evolutionary mechanisms in afrotherians ., Using the amino acid sequence of bornavirus N ( strain name: H1499 , accession number: AY374520 ) as the query , a tBLASTn search was conducted against whole genome shotgun sequences of afrotherians ( taxid: 311790 ) on June 24th , 2015 ., A total of 25 EBLNs were identified from African elephant ( Loxodonta africana ) , rock hyrax ( Procavia capensi ) , Florida manatee ( Trichechus manatus latirostris ) , Cape golden mole ( Chrysochloris asiatica ) , Cape elephant shrew ( Elephantulus edwardii ) , and aardvark ( Orycteropus afer ) , with an e-value threshold of E-20 ( S1 Table ) ., Two EBLNs each in African elephant , rock hyrax , Florida manatee , Cape elephant shrew , and aardvark contained relatively long ORFs encoding 341–350 aa , which were similar in length to bornavirus N ( 370 aa ) ( S1 Fig ) ., Although two EBLNs were identified in Cape golden mole , the ORF of one EBLN ( Ca/AMDV01031468 ) was separated into two fragments encoding 196 aa and 142 aa ., The other EBLN ( Ca/AMDV01100225 ) contained an ORF encoding 314 aa , which showed the highest amino acid sequence identity of 53% to bornavirus N among the afrotherian EBLNs identified in this study ., In the phylogenetic tree of afrotherian EBLNs and bornavirus N ( Fig 1A and S2 Fig ) , all EBLNs encoding 341–350 aa as well as the EBLN of Cape golden mole with separated ORFs ( Ca/AMDV01031468 ) formed a monophyletic cluster ( cluster I ) ., The topology of cluster I EBLNs in Fig 1A reflected that of the host species 21 ., In addition , the flanking genomic nucleotide sequences of cluster I EBLNs were alignable between afrotherians ( S3 Fig ) , suggesting that cluster I EBLNs originated from a single integration event of viral N gene into the genome of ancestral afrotherians more than 83 . 3 MYA ( Fig 1A ) 20 ., Afrotherian EBLNs with fragmented ORFs formed another cluster ( cluster II ) , which contained EBLNs from African elephant , Folorida manatee , and aardvark ., When EBLN sequences identified from non-afrotherian species were added to the phylogenetic tree , cluster I EBLNs and cluster II EBLNs were still monophyletic and constituted a larger cluster with Strepsirrhini EBLNs ( S4 Fig ) ., These results suggest that afrotherians have suffered from infection by bornavirus-like virus from the time of their origin and that colonizaiton of EBLNs has occurred at least twice during evolution of afrotherians ., When the degree of sequence divergence ( branch lengths ) was compared between cluster I and cluster II EBLNs , it was evident that the amino acid sequences encoded by the former EBLNs evolved more slowly than those encoded by the latter ( Fig 1A and S2 Fig ) , suggesting the functionality of amino acid sequences encoded by cluster I EBLNs ., Indeed , the average dN/dS ratio was estimated to be 0 . 34 and negative selection was detected ( p = 2 . 1 × 10−24 by the likelihood ratio test ) for the entire phylogenetic tree of cluster I EBLNs ( Fig 1C ) ., In addition , when the dN/dS ratio was estimated at each branch of the phylogenetic tree for cluster I EBLNs , the dN/dS ratio was smaller than one and negative selection was detected ( p < 0 . 05 by the likelihood ratio test ) at most of the branches ( Fig 1C and S2 Table ) ., These results indicate that cluster I EBLNs may encode functional proteins in afrotherians ., In cluster I , there were two EBLNs obtained from African elephant , named laEBLN-1 ( contig accession number: AAGU03015684 ) and laEBLN-2 ( contig accession number: AAGU03015682 ) ( Fig 1A ) ., These EBLNs contained ORFs encoding 344 aa of identical sequences except for a single site ( S1 Fig ) , each of which was 33% identical with the sequence of bornavirus N ( S5 Fig ) ., laEBLN-1 and laEBLN-2 were mapped ~45 kb apart on the same chromosome of African elephant ( Loxafr3 . 0 , supercontig 5 ) ( Fig 2 and S3 Table ) ., The genomic region encompassing these loci were flanked by transmembrane protein 106B ( TMEM106B ) ( accession number: XM_003407108 ) and scinderin ( SCIN ) ( accession number: XM_003407107 ) ., The syntenic relationship among TMEM106B , two copies of cluster I EBLNs , and SCIN as above was also observed in the genome of Florida manatee ( S6 Fig ) ., Although the syntenic relationship of TMEM106B and SCIN was observed in the genomes of many vertebrates ( S4 Table ) , no sequence element related to laEBLN-1 or laEBLN-2 was identified between them in non-afrotherian genomes ( S7 Fig ) , supporting the notion that the integration event of cluster I EBLNs took place on the lineage of afrotherians ., To examine whether laEBLN-1 and laEBLN-2 were transcribed into mRNAs , RT-PCR was conducted with a set of primers that was designed to amplify DNA fragments of the same size from these EBLNs ( S8 Fig and S5 Table ) ., A DNA fragment of the expected size was amplified from total RNA extracted from liver and muscle tissues from an adult male Asian elephant ( Elephas maximus ) and cell lines established from African elephant ear ( LACF-NANAI ) and gum ( LACF-NANAII ) tissues ( Fig 3A and S9 Fig ) ., These results suggest that mRNAs containing laEBLN-1 and/or laEBLN-2 and their orthologues are transcribed ubiquitously in African and Asian elephants , respectively ., Complete nucleotide sequences of mRNAs containing laEBLN-1 and laEBLN-2 , named laEBLN-1v1 ( accrssion number: LC093509 ) and laEBLN-2v1 ( accession number: LC093510 ) , respectively , were determined by 5’ and 3’ RACEs ., Both mRNAs consisted of two exons , 1 and 2 , where EBLN was embedded in exon, 2 . Exon 1 and its upstream region , exon 2 and its downstream region , and the intron were all homologous between the genomic loci for laEBLN-1v1 and laEBLN-2v1 ( S10 Fig ) , suggesting that these loci were generated through gene duplication at the DNA level ., It should be noted that mRNAs apparently transcribed from the same genomic loci as laEBLN-1v1 and laEBLN-2v1 but processed in alternative splicing forms have been deposited in the RNAseq database for African elephant ( BioSample: SAMN02953622 ) , and were named laEBLN-1v2 ( accession number: LOC104845604 ) and laEBLN-2v2 ( accession number: LOC104845603 ) , respectively , in this study ( Fig 2 ) ., Both of the alternatively spliced forms consisted of three exons , 1 , 2 , and, 3 . Exon 1 and a part of exon 2 were shared between laEBLN-1v1 and laEBLN-1v2 and between laEBLN-2v1 and laEBLN-2v2 ( S11 Fig ) ., When the mRNA expression of laEBLN-1v1 , laEBLN-1v2 , laEBLN-2v1 , and laEBLN-2v2 was examined separately by RT-PCR with four sets of primers that were designed to amplify individual mRNAs ( S8 Fig and S5 Table ) , laEBLN-1v1 , laEBLN-1v2 , and laEBLN-2v1 were detected in both LACF-NANAI and LACF-NANAII ( Fig 3B ) ., In contrast , laEBLN-2v2 was not detected in either cell line , suggesting a differentiation in expression patterns between splice variants ., The expected molecular weights of the proteins encoded by laEBLN-1v1 , laEBLN-1v2 , laEBLN-2v1 , and laEBLN-2v2 were 38 , 888 Da , 39 , 670 Da , 38 , 860 Da , and 39 , 642 Da , respectively ., To examine whether proteins were expressed from these mRNAs , rabbits were immunized with a recombinant laEBLN-1v1 ( rlaEBLN-1v1 ) protein expressed in E . coli to induce polyclonal antibodies , which were confirmed to react with rlaEBLN-1v1 ( Fig 4 , lane 6 ) ., The rabbit polyclonal antiserum was also found to react with both of laEBLN-1v1 and laEBLN-1v2 proteins expressed in 293F cells in the western blot analysis ( Fig 4 , lanes 3 and 4 ) ., When the western blot analysis using the rabbit polyclonal antiserum was performed on the whole protein extracts from LACF-NANAI and LACF-NANAII , a single band was observed at ~38 kDa ( Fig 4 , lanes 1 and 2 ) , indicating protein expression from some laEBLN mRNAs in African elephant cells ., To examine the cellular localization of each variant protein , FLAG-tagged laEBLN-1v1 and His-tagged laEBLN-1v2 proteins , which were confirmed to react with rabbit polyclonal antibodies ( Fig 4 , lanes 7 and 8 ) , were expressed in LACF-NANAI , and were stained with anti-DDDDK- and anti-His-mouse monoclonal antibodies ., Note that in the bioinformatic analysis of the amino acid sequences of laEBLN-1v1 and laEBLN-1v2 proteins using PSORT II 22 , the former was predicted to be localized to the cytoplasm , whereas the latter to the nucleus ( S6 Table ) ., Consistently , the signal of FLAG-tagged laEBLN-1v1 protein was detected in the cytoplasm ( Fig 5A ) , whereas the singal of His-tagged laEBLN-1v2 protein in the nucleus ( Fig 5B ) ., Similar cellular localizations of laEBLN proteins were also observed in 293F cells ( S12 Fig ) , suggesting that laEBLN-1v1 and laEBLN-1v2 proteins may play different functions ., It should be noted that bornavirus N is known to be localized to the nucleus of infected cells ., It is therefore conceivable that the protein product of laEBLN-1v1 may have acquired a novel function in afrotherians ., We then stained LACF-NANAI and LACF-NANAII with rabbit polyclonal antibodies to identify the cellular localizations of endogenous protein products from laEBLN mRNAs ., Positive signals were detected around the nucleus in the cytoplasm , co-localizing with the perinuclear part of the endoplasmic reticulum ( ER ) ( Fig 6A ) and , in particular , with the ribosome ( Fig 6B ) , suggesting that laEBLN proteins were associated with the rough ER ( rER ) in African elephant cells ., The presence and absence of positive signals in the cytoplasm and nucleus , respectively , may reflect relative abundance of laEBLN proteins in African elephant cells ., Assuming that the rabbit polyclonal antibodies could recognize all of the protein products from laEBLN-1v1 , laEBLN-1v2 , laEBLN-2v1 , and laEBLN-2v2 and the cellular localizations of the protein products from laEBLN-1v1 and laEBLN-2v1 were cytoplasmic and those from laEBLN-1v2 and laEBLN-2v2 were nuclear , the observed pattern of positive signals may be consistent with the result obtained above that mRNA expression of laEBLN-2v2 was not detected in African elephant cells ( Fig 3B ) ., In the genomic sequence of African elephant , laEBLN-1 and laEBLN-2 were both followed by transcription stop site ( T1 ) and polyA ( Fig 7 ) ., The sequence regions encompassing laEBLN-1 and its polyA and laEBLN-2 and its polyA were both flanked by TSDs with similar sequences ., In addition , a 6 nt sequence related to a transcription start site ( S1 ) of bornavirus was observed immediately downstream of the 5’ TSD ( TSD1 ) and 5 nt upstream of the start codon in laEBLN-1 ., These observations indicate that laEBLN-1 and laEBLN-2 have originated from a common integration event of a reverse-transcribed mRNA for viral N gene through the LINE-1 machinery , similarly to the case for the EBLNs previously identified in other animal species 8 , 10 ., Here it should be noted that bornavirus N mRNA does not contain an eukaryotic promoter sequence , and thus it is unclear how the integrated sequence element gained the ability to be transcribed in the host cell ., Interestingly , exon 1 and its flanking regions in the genomic loci for laEBLN-1v1/1v2 and laEBLN-2v1/2v2 in African elephant were discovered to be homologous to those for TMEM106B ( Fig 8A and S13 Fig ) ., In particular , the 5’ splice site ( GU ) for the first intron of laEBLN-1v1/1v2 and laEBLN-2v1/2v2 appeared to be derived from the corresponding site of TMEM106B ( S14 Fig ) ., It was then hypothesized that laEBLN-1 and laEBLN-2 gained the ability to be transcribed in the host cell by capturing a partial duplicate of TMEM106B , which contained a copy of the promoter and transcription start site ( TSS ) for TMEM106B ., To test this hypothesis , we conducted a promoter assay by constructing a series of luciferase expression plasmids , in which the luciferase gene was placed downstream of ( 1 ) exon 1 of laEBLN-1v1/1v2 ( pGL4 . 10E-1u ) , ( 2 ) exon 1 of laEBLN-2v1/2v2 ( pGL4 . 10E-2 ) , ( 3 ) exon 1 and upstream 454 nt of laEBLN-1v1/1v2 ( pGL4 . 10E-1u ) , and ( 4 ) exon 1 and upstream 442 nt of laEBLN-2v1/2v2 ( pGL4 . 10E-2u ) ( Fig 8B ) ., Note that 454 nt and 442 nt upstream of exon 1 of laEBLN-1v1/1v2 and laEBLN-2v1/2v2 , respectively , were homologous to the upstream sequence of exon 1 of TMEM106B ( Fig 8A ) ., It was observed that luciferase activities of pGL4 . 10E-1u and pGL4 . 10E-2u were ~100 times higher than those of pGL4 . 10E-1 and pGL4 . 10E-2 ( p < 0 . 01 by Student’s t test ) , respectively ( Fig 8B ) , supporting the above hypothesis ., In this study , we found that afrotherian EBLNs were clustered into two phylogenetically distinct classes , i . e . , cluster I and cluster II EBLNs , with an exception of the EBLN of Cape golden mole encoding 314 aa ., Cluster I EBLNs originated from a single integration event of N mRNA from a bornavirus-like virus into the ancestral genome of afrotherians through the LINE-1 machinery more than 83 . 3 MYA , which overlaps with the time period when LINE-1 was active in afrotherians 23–25 ., On the other hand , cluster II EBLNs were observed only in the genomes of African elephant , Florida manatee , and aardvark , suggesting that the integration event of cluster II EBLNs into the afrotherian genomes may have taken place relatively recently compared to cluster I EBLNs ., Amino acid sequences encoded by relatively long ORFs in cluster I EBLNs have been negatively selected , suggesting that they were co-opted in afrotherians as functional proteins ., In contrast , ORFs in cluster II EBLNs were fragmented and apparently have evolved without functional constraint at the amino acid sequence level ., The difference in the fates of cluster I and cluster II EBLN ORFs may have stemmed from the presence or absence of a partial duplicate of TMEM106B upstream of EBLNs in the genome ., The promoter and TSS of human TMEM106B are located in the CpG island ( S15 Fig ) , which is known to be associated with ubiquitously expressed genes such as house keeping genes , and human TMEM106B mRNA has been reported to be expressed ubiquitously 26–29 ., In the genome of African elephant , GC content in the 100 nt upstream of laEBLN-1v1/1v2 and laEBLN-2v1/2v2 is 65% and 61% , respectively , suggesting that the promoter and TSS of these mRNAs are also located in the CpG island ., Indeed , the overall expression of laEBLN-1v1/1v2 and laEBLN-2v1/2v2 was ubiquitous in elephant ., It is conceivable that the partial duplicate of TMEM106B provided cluster I EBLNs with an opportunity to be transcribed ubiquitously in afrotherians , which may have facilitated the EBLN proteins to acquire novel functions in the host before the occurrence of deleterious mutations in the ORF ., It should be noted , however , that acquisition of intrinsic promoter and TSS may not be necessary for transcription of EBLNs in the host cell , because all of seven EBLNs in human were shown to be transcribed in some tissues although their association with intrinsic promoters and TSS has not been identified 10 , 30 , 31 ., In addition , mRNAs containing the cluster II African elephant EBLN ( La/AAGU03023585 , RNAseq accession number: XM_010588552 ) and the Cape golden mole EBLN encoding 314 aa ( Ca/AMDV01100225 , RNAseq accession number: XM_006861214 ) were found in the RNAseq database ., It is possible that mRNAs containing these EBLNs were not expressed in sufficiently large numbers of tissues for acquisition of novel functions in the host before the occurrence of deleterious mutations in the ORF ., However , it should also be noted that the protein function once acquired by EBLNs can be lost during evolution of the hosts ., In afrotherians , the ORF of cluster I EBLN in Cape golden mole ( Ca/AMDV0103146 ) was separated into two fragments , and Lesser hedgehog tenrec ( Echionps telfairi ) contained only remnants of cluster I EBLNs ( contig accession numbers: AAIY02040498 and AAIY02084943 ) , which could not be detected by the tBLASTn search conducted in this study ., Interestingly , these species are closely related as members of Tenrecoidea ., These information may be useful for understanding the protein function of cluster I EBLNs in other species ., Cluster I EBLNs were tandemly duplicated in afrotherian genomes , and were transcribed in alternatively spliced forms in African elephant , generating laEBLN-1v1 , laEBLN-1v2 , laEBLN-2v1 , and laEBLN-2v2 ., Although these mRNAs as a whole appeared to be expressed ubiquitously , the expression profile of laEBLN-2v2 was diversified from those of laEBLN-1v1 , laEBLN-1v2 , and laEBLN-2v1 ., In addition , protein products of laEBLN-1v1 and laEBLN-2v1 were expected to be different from those of laEBLN-1v2 and laEBLN-2v2 at the C-terminal 20 aa , and the FLAG-tagged laEBLN-1v1 and His-tagged laEBLN-1v2 proteins showed different subcellular localizations in elephant cells ., These observations were indicative of a functional differentiation among laEBLN proteins ., Nevertheless , the splice donor site ( GU ) for the second intron to generate laEBLN-1v2 and laEBLN-2v2 were identified only in elephant , manatee , and hylax ( Fig 9A ) , suggesting that expression of splice variants corresponding to laEBLN-1v2 and laEBLN-2v2 may be limited to these species ., It was unclear whether these species gained or other species lost the splice variants because these scenarios were equally likely from the inference of ancestral sequences at the splice donor site according to the parsimony principle ( Fig 9B ) ., It has been reported that the protein product of squirrel EBLN ( itEBLN ) , which was identical to bornavirus N at 77% of amino acid sites , may suppress bornavirus infection by disrupting the function of N 19 ., In contrast , bornavirus infection was not suppressed by the protein product of hsEBLN-1 , which shared 41% of amino acid sequence with bornavirus N . The amino acid sequences encoded by afrotherian EBLNs were more divergent from bornavirus N than the sequence encoded by hsEBLN-1 ., In addition , endogenous laEBLN protein was observed to be localized to the rER in African elephant cells , which was in sharp contrast to the fact that bornavirus N is localized to the nucleus in infected cells 32 , 33 ., These observations suggest that laEBLN proteins have acquired a novel function associated with rER in afrotherian cells ., The rER is associated with ribosomes and involved in the translation of cytoplasmic , secretory , and membrane proteins ., There are mechanisms to deliver mRNAs from the cytoplasm to the rER membrane by the action of RNA-binding proteins , such as STAU1 , STAU2 , Pum1 , and Pum2 34–36 ., Because mononegavirus N has an ability to bind to RNA and laEBLN proteins are localized to rER , it is interesting to assess the involvement of laEBLN proteins in mRNA delivery ., The hydropathy plot of the amino acid sequence encoded by laEBLN-1v1 showed that laEBLN-1v1 protein may be soluble ( Fig 10 ) , which was consistent with the characteristic of the rlaEBLN-1v1 protein produced in E . coli ( See Materials and Methods ) ., In the window analysis of the dN/dS ratio for the ORF in cluster I EBLNs , it appeared that negative selection has operated more strongly on hydrophilic regions ( average dN/dS ratio = 0 . 41 ) than on hydrophobic regions ( average dN/dS ratio = 0 . 48 ) ( p < 0 . 05 by z-test where the variance of average dN/dS ratio was estimated with bootstrap resampling of windows ) ( Fig 10 ) , suggesting that the protein product of cluster I EBLNs may interact with other molecules and the interaction may be critical in afrotherians ., It is interesting to clarify the function of laEBLN proteins to understand the impact of viruses on the evolution of their hosts ., Afrotherian EBLNs were identified by a tBLASTn search using the amino acid sequence of bornavirus N ( strain name: H1499; accession number: AY374520 ) as the query against the database of whole genome shotgun ( WGS ) sequences for afrotherians ( taxid: 311790 ) on June 24th , 2015 ., Sequence hits with an e-value threshold of E-20 were identified as EBLNs ., EBLNs in non-afrotherian species were also identified from the database of WGS sequences for vertebrates ( taxid: 7742 ) on May 18th , 2016 by the same criterion as described above ., Amino acid sequences encoded by EBLNs were subjected to HMMER for examining the existence of domains that have been deposited in the Pfam database ., Multiple alignments of amino acid sequences for EBLNs and bornavirus N were made by mapping each of the amino acid sequences encoded by EBLNs onto that of bornavirus N ( strain name: H1499; accession number: AY374520 ) according to the pairwise alignment of these sequences produced in the tBLASTn search ( S1 File ) ., In each EBLN sequence , the amino acid sites differentially aligned to bornavirus N in different tBLASTn hits were treated as gaps ., In a dot-plot analysis , the genomic sequences of the genes of interest together with their upstream 1 , 500 nt were retrieved from Ensemble on 12th May , 2015 , and were compared using YASS 37 ., Genomic sequences of the regions 46 , 200 , 000–46 , 700 , 000 in scaffold_5 , supercontig loxFar3 of African elephant and 12 , 200 , 000–12 , 700 , 000 in chromosome 7 , GRCh38 of human were subjected to synteny analysis using GeneMatcher ( version 2 . 014 ) 38 , in which BLASTn and tBLASTn searches were conducted for identifying homologous segments between these sequences ., The phylogenetic tree of EBLNs and bornavirus N was constructed by the maximum likelihood ( ML ) and nighbour-joing ( NJ ) methods with the partial deletion and pairwise deletion options in MEGA 6 , respectively 39 ., The JTT+G model was chosen as the best fit model of amino acid substitution with the smallest Bayesian information criterion score ., The nearest-neighbor interchange was selected as the ML heustric method ., The reliability of interior branches in the phylogenetic tree was assessed by computing the bootstrap probability with 1 , 000 resamplings ., Divergence times between afrotherians were obtained from TimeTree 20 ., The dN/dS ratio for the entire phylogenetic tree as well as for each branch of cluster I EBLNs was estimated by the ML method using the codon substitution model in PAML ver . 4 . 0 40 ., The equilibrium codon frequencies were treated as free parameters ., The selective neutrality was tested for the entire phylogenetic tree or for each branch by the likelihood ratio test ., Window analysis of the dN/dS ratio was conducted between laEBLN-1v1 and each of other cluster I EBLNs , i . e . , Tm/AHIN01138425 , Tm/AHIN01138426 , Pc/ABRQ02082163 , Pc/ABRQ02082168 , Oa/ALYB01141940 , Oa/ALYB01141942 , Ee/AMGZ01016176 , and Ee/AMGZ01016178 , with a window size of 20 codons and a step size of 1 codon using ADAPTSITE 41 ., Hydropathy scores along the amino acid sequences encoded by laEBLN-1v1 , Tm/AHIN01138425 , Tm/AHIN01138426 , Pc/ABRQ02082163 , Pc/ABRQ02082168 , Oa/ALYB01141940 , Oa/ALYB01141942 , Ee/AMGZ01016176 , and Ee/AMGZ01016178 were calculated as the Kyte and Doolittle index with a window size of 20 amino acids and a step size of 1 amino acid using GENETYX ( version 10 . 1 . 1 ) ( Genetyx , Tokyo , Japan ) ., Liver and muscle tissue samples were collected from an adult male of Asian elephant dead at Kobe Oji Zoo in Japan , and were stored at -80°C until use ., Cell lines LACF-NANAI ( RIKEN Cell Bank: RCB2319 ) and LACF-NANAII ( RIKEN Cell Bank: RCB2320 ) , which had been established from the gum and the ear of African elephant , respectively , were provided by RIKEN Cell Bank , Japan , and were maintained in Dulbecco’s modified Eagle’s medium ( DMEM; GIBCO/BRL ) containing 10% fetal bovine serum ( FBS ) , L-glutamine , and penicillin-streptomycin under 5% CO2 at 37°C ., The 293F cells ( Invitrogen ) were maintained in Eagle’s minimum essential medium ( EMEM; GIBCO/BRL ) containing 5% FBS , L-glutamine , and penicillin-streptomycin under 5% CO2 at 37°C ., BHK-21 cells ( RIKEN Cell Bank: RCB1423 ) , which were provided by RIKEN Cell Bank , were maintained in DMEM containing 10% FBS , L-glutamine , and penicillin-streptomycin under 5% CO2 at 37°C ., Total RNA was extracted from the liver and muscle tissues of an Asian elephant; 50 mg of each tissue was frozen with beads ( TOMY ) and 1 ml of ISOGEN ( Nippon Gene ) in liquid nitrogen , and homogenized using TOMY Micro Smash MS-100R ( TOMY ) ., The homogenized sample was mixed with 200 μl of 100% chloroform , and the mixture was centrifuged at 12 , 000 × g for 15 min at 4°C ., The supernatant with 600 μl of 70% ethanol was added to an RNeasy Spin Column provided in the RNeasy Mini Kit ( QIAGEN ) , and RNA was extracted following manufacturer’s instructions ., Total RNA was also extracted from LACF-NANAI , LACE-NANAII , Baby Hamster Kidney ( BHK ) cells , and 293F cells using an RNeasy Mini Kit according to manufacturer’s instructions ., DNase digestion was performed on all samples during the extraction process using RNase-free DNase set ( QIAGEN ) ., Expression of mRNAs containing laEBLN-1 and laEBLN-2 or their orthologues in the elephant samples was examined by RT-PCR ., One-step RT-PCR was conducted using a SuperScript III/Platinum Taq one-step RT-PCR kit ( Invitrogen ) in a final volume of 25 μl containing 1 × reaction mixture , 0 . 4 μM primers , 1 μl SuperScript III RT/Platinum Taq Mix , and 20 ng of total RNA ., Primer sequences were designed with Primer Blast in NCBI as listed in S5 Table ., RT-PCR was performed as follows; reverse transcription at 50°C for 60 min , 40 cycles of 94°C for 30 sec , 60°C for 30 sec , and 72°C for 30–90 sec , followed by a final extension at 72°C for 3 min ., In 2-step RT-PCR , cDNA was synthesized from mRNAs using oligodT primer with or without SuperScript III Reverse Transcriptase ( Invitrogen ) , and the product was used as the template for PCR ., In brief , PCR was performed in a final volume of 25 μl containing 1 × PCR Buffer , 0 . 2 mM dNTP , 0 . 4 μM primers , 1 . 25 U Blend Taq ( TOYOBO ) , and 1μl of the above template ., PCR reaction was performed as follows; 94°C for 2 min and 40 cycles of 94°C for 30 sec , 55°C for 30 sec , and 72°C for 30 sec ., The size of the products was analyzed by agarose gel electrophoresis ., The 5’ and 3’ RACEs for laEBLN-1v1 and laEBLN-2v1 were performed using 2 . 75–3 . 75 μl of total RNA collected from LACF-NANAII with SMARTer RACE cDNA Amplification Kit ( Clontech ) ., Briefly , after incubation of the total RNA with 12 μM 5’ and 3’ CDS primers at 72°C for 3 min and at 42°C for 2 min , the first-strand cDNA synthesis for 3’ ( 5’ ) RACE was performed in a final volume of 10 μl containing 1 × First-Strand Buffer , 20 mM DTT , 10 mM dNTP mix , 10 U RNase Inhibitor , ( 12 μM SMARTer II A Oligonucleotide , ) and 50 U SMARTScribe Reverse Transcriptase at 42°C for 90 min , followed by a termination process at 70°C for 10 min ., The reaction mixture was diluted to 100 μl with Tricine-EDTA buffer , and was used as the template for PCR in a final volume of 25 μl containing 1 × PCR Buffer , 0 . 2 mM dNTP , 0 . 4 μM primers , 1 . 25 U Blend Taq ( TOYOBO ) , and 1 μl of the above cDNA mixture ., Nested-PCR was performed using the PCR products diluted 500-fold with distilled water , and the products were purified with Wizard SV Gel and PCR Clean-Up System ( Promega ) ., The nested-PCR products were ligated into the pGEM-T Easy vector ( Promega ) using T4 DNA ligase ( New England Biolabs ) , and the plasmids were transformed into TOP10 Competent Cells ( Life Technologies ) ., For each of 5’ and 3’ RACEs , at least 10 colonies were selected by direct-colony PCR and the plasmids were purified from the colonies using the Wizard Plus SV Minipreps DNA Purification System ( Promega ) ., Nucleotide sequences of the inserts were determined using BigDye Terminator v3 . 1 Cycle Sequencing Kit with ABI Prism 3130 ( ABI ) ., Genomic DNA was extracted from LACF-NANAII using a QIAamp DNA Blood Mini Kit ( QIAGEN ) according to manufacturer’s instructions ., The ORF of laEBLN-1v1 was amplified from genomic DNA with the primer pair AFEBLN_pET_F and AFEBLN_pET_R ( S5 Table ) ., The PCR product was ligated into the PET100 vector ( Invitrogen ) using Ligation high ver . 2 ( TOYOBO ) at 16°C for 30 min , and the plasmid was transformed into BL21 cells ., Colonies harboring the plasmids were selected and propagated in LB medium containing 0 . 02% lactose , 0 . 05% glucose , 0 . 5% glycerine , 2 mM MgSO4 , and phosphate buffer for overexpression of the protein encoded by laEBLN-1v1 ., BL21 cells were disrupted using ultrasonic wave , and centrifuged at 6 , 000 × g for 30 min at 4°C ., The recombinant laEBLN-1v1 ( rlaEBLN-1v1 ) protein was purified from the supernatant with His-Trap HP and Hi-trap desalting column ( GE Healthcare ) , and was used as the antigen for immune induction ., In the

Introduction, Results, Discussion, Materials and Methods

Endogenous bornavirus-like nucleoprotein elements ( EBLNs ) , the nucleotide sequence elements derived from the nucleoprotein gene of ancient bornavirus-like viruses , have been identified in many animal genomes ., Here we show evidence that EBLNs encode functional proteins in their host ., Some afrotherian EBLNs were observed to have been maintained for more than 83 . 3 million years under negative selection ., Splice variants were expressed from the genomic loci of EBLNs in elephant , and some were translated into proteins ., The EBLN proteins appeared to be localized to the rough endoplasmic reticulum in African elephant cells , in contrast to the nuclear localization of bornavirus N . These observations suggest that afrotherian EBLNs have acquired a novel function in their host ., Interestingly , genomic sequences of the first exon and its flanking regions in these EBLN loci were homologous to those of transmembrane protein 106B ( TMEM106B ) ., The upstream region of the first exon in the EBLN loci exhibited a promoter activity , suggesting that the ability of these EBLNs to be transcribed in the host cell was gained through capturing a partial duplicate of TMEM106B ., In conclusion , our results strongly support for exaptation of EBLNs to encode host proteins in afrotherians .

Endogenous retroviruses are representative of endogenous viral elements ( EVEs ) , which are known to have occasionally served as the source of evolutionary innovations of the host ., Endogenous bornavirus-like nucleoprotein element ( EBLN ) was the first EVE identified in mammalian genomes to have been derived from a non-retroviral RNA virus ., Here we show evidence that EBLNs that were integrated into afrotherian genomes more than 83 . 3 million years ago have gained novel protein functions associated with rough endoplasmic reticulum in afrotherians ., In the amino acid sequence of EBLN proteins , negative selection appeared to have operated more strongly on hydrophilic regions than on hydrophobic regions , suggesting that EBLN proteins may interact with other molecules in their host cells ., In addition , we clarified the mechanism how EBLNs have acquired an ability to be transcribed in the host cell; they captured a partial duplicate of an intrinsic gene , transmembrane protein 106B , which retained an intrinsic promoter activity ., Our findings suggest that not only retroviral EVEs but also non-retroviral EVEs may have contributed to the host evolution .

sequencing techniques, split-decomposition method, messenger rna, vertebrates, animals, mammals, plasmid construction, viruses, multiple alignment calculation, rna viruses, bornaviruses, dna construction, molecular biology techniques, research and analysis methods, sequence analysis, sequence alignment, artificial gene amplification and extension, molecular biology, elephants, biochemistry, rna, computational techniques, nucleic acids, polymerase chain reaction, biology and life sciences, amniotes, protein sequencing, organisms

journal.pgen.0030034

A Caenorhabditis elegans Wild Type Defies the Temperature–Size Rule Owing to a Single Nucleotide Polymorphism in tra-3

For many decades biologists have been intrigued by the relation between body size and temperature ., It was discovered that ectotherms—animals that maintain their body temperature by absorbing heat from the surrounding environment such as fish and all invertebrates—reproduce later at a larger size when reared at lower temperatures 1–3 ., This phenomenon is known as the temperature–size rule , and nearly 90% of ectothermic species studied so far follow this rule 4 ., The magnitude of this phenomenon is illustrated by Azevedo et al . 5 who found a 12% increase in wing and thorax size in Drosophila melanogaster when grown at relatively low temperatures ., In the case of the nematode C . elegans ( strain Bristol N2 ) , an environmental temperature of 10 °C resulted in adults that were ~33% larger than those grown at 25 °C 6 ., About 99 . 9% of all species are ectothermic , and the temperature–size rule is observed in bacteria , protists , plants , and animals , making it one of the most widespread phenomena in ecology ., From the perspective of life-history evolution it is not well understood why growing bigger at lower temperatures is beneficial for organisms ., Because this thermal plasticity of body size is taxonomically widespread , the reasons are probably diverse and may vary among groups of organisms ., It has been suggested that a large body size is advantageous , because it compensates for delayed reproduction by yielding more offspring 7 ., Other explanations may be that a larger body size at maturity enables individuals to produce larger offspring or to provide better parental care 2 ., Since body size and temperature are the two most important variables affecting fitness 8 , 9 , many experimental and theoretical attempts have been made to explain the mechanism underlying the temperature–size rule ., Essentially , an increase in body size can be achieved by increasing cell number , cell size , or by both ., Various studies point at the second ( cell size ) and the third option ( cell size and number ) as being the most likely explanation for the observed increase in body size at lower temperatures ( Drosophila spp . 10–12 , yellow dung fly 13 , and the nematode C . elegans 6 ) ., Next to these empirical observations , various models have been proposed that are based on a combination of changes in cell size and number ., Biophysical models show that the temperature–size rule is the result of unequal effects of temperature on cell growth and cell division 14 ., When the effect of temperature on the rate of division is greater than its effect on the rate of cell growth , the model predicts that a low temperature should lead to a larger body size ., Recently , a physiological model was proposed by Atkinson et al . 15 , which assumes that temperature induced changes in cell size and number depend on the optimisation of oxygen supply at different temperatures ., Yet , these empirical and theoretical findings give little insight into the molecular genetic control of body size at lower temperatures ., Unravelling the molecular mechanism underlying the temperature–size rule is hampered by the fact that temperature affects nearly all biochemical processes in a cell , and in theory growing bigger at lower temperatures may have numerous causes ., However , low temperatures also have been shown to induce a number of specific physiological and genetic responses in ectotherms 16 ., In D . melanogaster gene expression analysis revealed a senescence marker smp-30 to be induced by low temperature 17 ., Van ‘t Land et al . 18 reported the association of the gene Hsr-omega with low temperatures in D . melanogaster ., Next to these specific gene responses , an early indicator of low temperature is a transient elevation of the cytosolic calcium concentration Ca2+i ., Higher cytosolic calcium levels occur not only in response to a rapid cooling but also to more gradual reductions in temperature , and it is a widespread phenomenon observed in plants 19 , 20 and ectothermic animals 21–24 ., Here we aimed to identify and characterize genes underlying the temperature–size rule in a model ectotherm , the nematode C . elegans ., C . elegans is a suitable model for studying the molecular control of temperature–body size responses because of its completely sequenced genome , isomorphic growth , and cell constancy , and because nematode life-history traits are easy to observe 25 ., We found that wild-type Bristol N2 ( designated as N2 ) grew bigger at lower temperatures and thus complied with the temperature–size rule , whereas wild-type CB4856 ( designated as CB ) defied the rule ., The natural variation in body size response to temperature between CB and N2 was caused by a single mutation F96L in a calpain-like protease TRA-3 encoded by tra-3 ., Homology modelling predicts that F96L is likely to reduce the ability of TRA-3 to bind calcium ., We studied the thermal reaction norm for body size ( TRB ) , which is a plot of body size at maturity versus temperature , and defined compliance with the temperature–size rule if body size is significantly and negatively related to temperature ., To assess differences in the TRB between the two wild-type strains we measured body size at 12 °C and 24 °C ., Body-size measurements were taken from Gutteling et al . 26 ., We found a marked difference in TRB between the two wild types ., The body size of wild-type N2 exhibited a significant negative relationship with temperature , i . e . , N2 grew larger at low temperature ( F = 3 . 49; p = 0 . 02 ) ., In contrast , CB defied the temperature–size rule because body size was not significantly affected by temperature ( F = 0 . 8; p = 0 . 47 ) ( Figure 1 ) ., The results for N2 are in agreement with previous findings where increased body size was found in C . elegans N2 hermaphrodites as well as males at lower temperatures 27 , 6 ., To further study the genetic control of the TRB , we first developed an N2 × CB recombinant inbred panel and performed a quantitative trait locus ( QTL ) analysis for detecting genomic regions associated with the TRB ., By selfing the CB × N2 F1 offspring for 20 generations , we obtained a segregating population of recombinant inbred lines ( RILs ) , which were also exposed to 12 °C and 24 °C ., We found large differences in TRB slopes among the RILs ( Figure 1 ) ., As generally observed in recombinant inbred crosses between divergent strains , the mean trait values for many of the RILs exceeded the mean value for either parental strain ., Apparently the differences between the N2 and CB phenotypes ( the slope of the TRB ) capture a great deal of genetic variation ., This was evident in the variation exhibited in the RILs for the TRB slope ., Such transgressive segregation has been reported for many organisms and indicates that alleles at different loci act in the same direction , and when combined these alleles will result in phenotypes more extreme than either parent 28 ., In general RILs matured at 12 °C at a bigger size than at 24 °C , which is in accordance with the temperature–size rule ( see Atkinson 4 for an overview ) ., We found strong genetic variation among RILs for body size across the two temperatures ( F = 40 . 1; p < 0 . 001 ) ., We then sought to determine which loci were associated with the TRB by genotyping the RILs and performing a QTL mapping study using the recombinant inbred panel ., For the QTL analysis we used a dense single nucleotide polymorphism ( SNP ) map ., A full description of the genetic architecture of the RILs can be found in 29 ., In summary , the overall average distance between two SNP markers was 835 kb or 2 . 38 cM ., The overall average chromosomal coverage was 96% if measured in bp or 95% if measured in cM ., Compared to the Wormbase F2-derived genetic maps ( http://www . wormbase . org , release WS106 ) , the genetic maps showed on average an ample 2-fold expansion ., This is common for RILs bred by self-fertilization or sib-mating and can be explained by the multiple rounds of meiosis undergone 30 ., Figure 2 shows the detected QTLs associated with the slope of the TRB ., Two QTLs on Chromosome IV were associated with a negative effect on the slope of the TRB and were linked to CB alleles ., The distal QTL at Chromosome IV showed pleiotropy or linkage for body size at 24 °C ( additive effect of 4% ) ., We aimed to identify the gene ( s ) controlling the QTL at Chromosome IV with a peak at marker pkP4095 at 12 cM , because this QTL was uniquely associated with TRB ( hence we named it the TRB-locus ) and not with body size itself at 12 °C or 24 °C ., This locus had a relatively large additive effect of 34% of the total standard deviation and explained 11% of the among-RIL variance ., Introgression of a CB segment spanning the TRB-locus into an N2 background confirmed the QTL analysis ., Phenotyping of NIL WN17–9 carrying an ~6-cM region of the TRB locus revealed no significant body-size difference between low and high temperature ( Figure 3 ) ., Three other QTLs on Chromosome III increased the slope and each of these QTLs was linked to N2 alleles and showed a pleiotropic or close linkage effect for body size at 12 °C 26 ., The 2 . 5-cM genome segment covered by the confidence interval ( CI ) of the TRB locus harbours a number of mutationally mapped genes of which only one ( dpy-4 ) 31 is known to affect body size ., To identify promising candidate genes , we reasoned as follows ., Previous studies have shown that body size in C . elegans is controlled by genes that affect cell size and not cell number 32 , 33 ., It is also known that this is one of the main mechanisms , next to cell number , by which ectotherms grow bigger in colder environments 7 ., Furthermore , we sought to identify and characterize genes that encode a calcium-activated protein because Ca2+i is a key signal of low temperature ., Lower temperatures lead to an increase of Ca2+i 21–24 ., Given these two facts ( increased Ca2+i and cell size ) we searched for genes that are activated by Ca2+i and that play a role in increased cell size ., Among the few genes with known function in the TRB locus , the most likely candidate gene was tra-3 ., TRA-3 has a high homology with mammalian calpains 34 , which are known to be activated by Ca2+i and have been reported to regulate cell size during oncosis ( cell swelling ) 35 ., dpy-4 is not known to be activated by Ca2+i 31 ., We therefore selected tra-3 as a candidate gene that might explain the difference in temperature responsiveness between N2 and CB ., The gene tra-3 seems to be important for the TRB slope because a significant difference was found between tra-3 allelic variants ( using the linked marker pkP4095 ) and the TRB slope ( t-test , p = 0 . 03 ) ., RILs with the N2 allele had a larger slope than RILs with a CB allele ., To investigate the hypothesis that tra-3 controlled the TRB , we first sequenced this region in CB ., One SNP was found within the coding region where phenylanaline-96 in N2 was mutated into leucine-96 in CB ., To see whether other tra-3 mutants displayed the same phenotype as observed in CB , we selected two homozygous artificial allelic mutants in an N2 background , tra-3 ( e1107 ) carrying a nonsense mutation 34 and tra-3 ( e2333 ) ., We also sequenced tra-3 ( e2333 ) in the ORF ± 1 kb and found a nonsense mutation at nucleotide position 1 , 779 ( G to A ) of the spliced tra-3 transcript ., This resulted in a premature stop ( W to stop ) at position 593 of the TRA-3 protein ., Both mutants were phenotyped for body size at 12 °C and 24 °C and compared to the wild-type N2 ., Like CB , body size was not affected by temperature in both mutants ( Figure 4 ) ., The N2 phenotype was rescued by the fully suppressed mutant tra-3 ( e1107 ) sup-24 ( st354 ) IV , which promotes translational readthrough of the tra-3 ( e1107 ) mutation ( Figure 4 ) ., We then tested whether a larger body size could also be obtained by mimicking a low temperature environment through an artificial increase of Ca2+i at 24 °C ., Although TRA-3 does not have a specific EF calcium-binding site in C . elegans , a well-conserved region has been shown to bind calcium 36 , 37 ., We used thapsigargin ( TG ) to increase Ca2+i 38 , 39 ., We found a clear dose–response relationship between TG and body size , showing that N2 grew larger at 24 °C at increased levels of TG ( Figure 4 ) ., A significant increase in size was found at 0 . 015 μM TG compared to a positive control that included the solvent dimethyl sulfoxide ( DMSO ) ., Calpain activity was required for the TG-induced body-size enlargement because treatment with 0 . 015 μM TG did not result in a larger body size in homozygous tra-3 ( 1107 ) mutants ( Figure 4 ) ., These results indicate that calcium activation of TRA-3 may be controlling body size at different temperatures ., In addition to the F96L mutation , the observed phenotypic differences could be due to differential expression of tra-3 ., Therefore , we performed quantitative RT-PCR experiments on cDNA obtained from N2 and CB at 12 °C and 24 °C ., It was found that expression was slightly enhanced at 24 °C in both wild types ., There was no significant difference in tra-3 expression across temperatures between N2 and CB ( results not shown ) ., Based on these findings we hypothesised that observed TRB differences between N2 and CB were the result of a polymorphism in tra-3 ., To further investigate the role of tra-3 in the wild-type CB , we performed a complementation analysis by crossing the near-isogenic line ( NIL WN17–9 ) with tra-3 ( e1107 ) ., Heterozygous F1 from a cross between NIL WN17–9 and N2 revealed the recessive nature of the CB–TRB allele ( Figure 3 ) ., The body size for the e ( 1107 ) /+ F1 offspring exhibited increased size at 12 °C indicating that tra-3 ( e1107 ) was recessive ( Figure 3 ) ., Complementation analysis in which NIL WN17–9 was crossed with tra-3 ( 1107 ) showed no differences in body size of F1 between high and low temperature ( Figure 3 ) ., These results show that tra-3 is required for regulating body size in response to changing environmental temperatures and that an SNP in tra-3 is able to reduce this ability ., We did not attempt to perform a complementation test between NIL WN17–9 and tra-3 ( e2333 ) because of the dominant nature of tra-3 ( e2333 ) over other tra-3 mutants ., Homozygous tra-3 ( e1107 ) worms show partial masculinisation whereas homozygous tra-3 ( e2333 ) animals are wild-type hermaphrodites ., Heteroalleles of these two mutants are also wild-type hermaphrodites indicating a dominance effect of e2333 over e1107 40 ., We next asked whether the N2 version of the tra-3 gene could transform CB to have a larger body size at low temperature ., Therefore we carried out a transgenic assay in which tra-3 from N2 was transferred to the CB background ., We exposed independently derived strains of CB ( gfp ) ( control strains ) and CB ( gfp and tra-3 ( + ) ) to 12 °C and 24 °C ., Figure 5 shows that the N2 phenotype was rescued in CB ( gfp and tra-3 ( + ) ) because it grew 24% larger at the low temperature ., CB ( gfp ) retained the CB phenotype because it did not grow larger at low temperature ., We next sought to determine whether F96L could lead to a diminished activity of TRA-3 in CB by conducting homology modelling of the 3D structure of TRA-3 ., The TRA-3 protein consists of four domains ( I–III and T ) , where domain II is the protease catalytic site , and domain T does not have a critical calcium-binding function 34 , 41 but may be important for protein folding ., Although TRA-3 does not have a specific EF calcium-binding site in C . elegans , a well-conserved region spanning the boundaries of domain II and III has been shown to bind calcium 36 ., In addition , Moldoveanu et al . 37 reported on non-EF calcium-binding sites in domain II between position 62–74 ( Ca-1 ) ., Homology modelling shows that F96 is located at the beginning of a short helix , H6 , contiguous in space to the loop hosting Ca-1 ., In “open” configuration , corresponding to the absence of calcium , the distance between the α-carbons of F96 and E68 , G69 , and A70 reduces to 8–10 Å , as compared to ~10–14 Å corresponding to the “closed” configuration ., In addition , the side chain of F96 is oriented toward the Ca-1 loop making their atoms to come frequently in van der Waals contact ( <3 . 0 Å ) ( Figure 6 ) ., As the length of a leucine side chain is ~1 . 5 Å smaller than that of a phenylalanine , F96L will introduce a void in this region ., Therefore , F96L can make a small but important difference by increasing the conformational space that the “opened” Ca-1 loop can sample during its dynamics ., As the number of configurations increases this might reduce the probability to find the loop in its “closed” configuration and consequently reduce the ability for calcium binding ., The genetic control of the C . elegans body size has been intensively studied ., Mutants such as sma-2 , 3 , 4 , and daf-4 have a small body size and are defective in the TGF-β signalling pathway , which underlies body growth and development 42 ., The lon mutants have been found to grow longer but not larger in volume 32 , 43 ., It was shown that egl-4 mutants , defective in a gene encoding a cGMP-dependent protein kinase , have a much larger body size than N2 32 ., Here it is shown that TRA-3 has a prominent role in regulating the thermal plasticity of body size in C . elegans ., Homology modelling shows that the F96L mutation in CB4856 attenuates the ability to grow bigger at lower temperatures by destabilizing the calcium-binding site in TRA-3 ., These data indicate that calcium signalling in response to temperature changes may lead to the activation of TRA-3 ., This mechanism to control the temperature–size rule is supported by various reports on the elevation of the free cytosolic calcium concentration in response to lower temperatures ., Increase of cytosolic calcium levels in response to a gradual reduction of temperature is widely observed in plants 19 , 20 and ectothermic animals ., Many studies in other organisms have shown the importance of calpains in oncosis showing calpain-mediated cell swelling and disruption of plasma membrane permeability followed by cell death 35 ., In C . elegans calcium-activated TRA-3 is known to be involved in the sex determination pathway by activating TRA-2A , a membrane protein that indirectly activates the zinc finger transcriptional regulator TRA-1A by binding and inhibiting a masculinising protein FEM-3 44 ., Current insights are insufficient to link aforementioned findings and to infer a putative pathway by which calcium activation of TRA-3 results in larger cell sizes in C . elegans ., Many different theories have been proposed to unravel the underlying mechanism of the temperature–size rule 2 , 7 , 45 , 46 ., Our results partly fit the theory by Van der Have et al . 14 who suggested that the temperature–size rule is regulated by two distinct processes underlying temperature effects on body size: growth rate ( which is the biomass increase per time unit ) and differentiation rate ( which is the reciprocal of development time ) ., Their model presupposes that the temperature–size rule depends on a wide range of alleles differing in sensitivity to temperature ., Our results show that a polymorphism in a single gene may attenuate the TRB in C . elegans ., CB was originally isolated in Hawaii while N2 originates from the UK ., Whether the F96L mutation in CB reflects adaptive change or a fortuitous event is unknown ., Both parental strains have been isolated decades ago and kept in the laboratory ever since , and additional field research is needed to establish whether this polymorphism and/or others in tra-3 are typical for strains isolated from tropical regions ., Our results do not provide insight into how natural selection modifies the temperature–size rule , yet they provide the basis for a more mechanistic understanding of the evolutionary outcomes ., Like C . elegans the increased body size at lower temperatures in flatworms , Drosophila spp ., , and protists 47–49 is caused primarily by increased cell size ., Because tra-3 shows a high homology with other ectothermic calpains 34 , 37 , our findings may imply a possible role of calpain in the control of the temperature–size rule in other organisms as well ., We have presented genetic and structural evidence that an SNP in the gene tra-3 encoding a calpain-like protease is required for the regulation of the temperature–size rule in wild-type C . elegans ., First , we found that the wild-type N2 complied with the temperature–size rule , whereas wild-type CB4856 defied it , and demonstrated that the genetic variation in the temperature–size response mapped to a single QTL on Chromosome IV harbouring tra-3 ., Second , we showed similar expression levels in tra-3 between the two wild types ., Third , transgenic CB carrying an N2 allele of tra-3 complied with the temperature–size rule ., Fourth , we found that F96L in TRA-3 attenuates the ability of wild-type CB4856 to grow larger at low temperatures ., Finally , we showed that , based on homology modelling , the CB4856 mutation decreased the calcium-binding activity of TRA-3 rendering it less active ., Because TRA-3 shows a high homology with other ectothermic calpains , our findings imply a possible role of tra-3 in the control of the temperature–size rule in other organisms as well ., Together our data show that the response of a quantitative trait to temperature changes can be simple and far less complex than previously thought ., Both N2 and CB parental strains were homozygous ., Strains were grown in 9-cm petri dishes at 15 °C or 20 °C on standard nematode growth medium with Escherichia coli strain OP50 as food source 50 and transferred to new dishes by a chunk of agar once a week ., RILs were constructed according to 29 ., NIL WN17–9 was constructed by crossing a single L4 hermaphrodite of RIL WN17 with five males generated from N2 on a 6-cm petri dish ., The proportion of males in the offspring was approximately 0 . 5 indicating a successful cross ., Subsequently , 12 crosses were set up , each with the use of single L4 hermaphroditic offspring of the former cross and five males derived from N2 ., Backcrossing procedure was continued with two L4 hermaphroditic offspring per successful cross ., After described three generations backcrossing , six L4 hermaphrodites were picked from each successful cross and placed individually on a 3-cm petri dish to self ., Selfing was continued for ten generations for each of the lines ., All derived lines were subsequently genotyped at seven marker positions ( including marker pkP4095 ) distributed equally over the fragments that were identified in RIL WN17 to be of CB origin ., A total of five lines that appeared to have N2 alleles in all genotyped positions except for the marker pkP4095 were used for detailed genotyping ., These lines were genotyped at all remaining marker positions ., The results for one of the genotyped lines ( NIL WN17–9 ) showed at all genotyped markers N2 alleles except for CB allele at marker pkP4095 , and three neighbouring marker positions at Chromosome IV indicating a single ~6-cM DNA fragment of CB origin introgressed into N2 background ., Genotyping was according to Li et al . 29 ., Prior to an experiment , all lines ( 80 RILs and two parental ) were synchronised at room temperature by transferring five adult nematodes to fresh 6-cm petri dishes and allowing them to lay eggs for 3–4 h , after which the nematodes were removed ., Eggs were allowed to develop at 20 °C , and three days later synchronisation was repeated to get double-synchronised lines ., Measurements for parental and RIL body size at maturation were taken from Gutteling et al . 26 ., Maturation was defined as the moment that the first few eggs are laid and can be easily observed ., Because of this , body size at maturity can be precisely measured ., For each RIL , three replicate experiments were performed using double-synchronised lines as a start ., In each replicate , four adult nematodes per RIL were transferred to a fresh 6-cm dish , allowed to reproduce at room temperature for 2–4 h ( average 2 . 5 h ) , and removed ., Dishes were then stored at 12 °C and 24 °C climate chambers ( Elbanton , http://www . elbanton . nl ) ., Temperature was recorded with Tinytag Transit temperature loggers ( Gemini Data Loggers , http://www . geminidataloggers . com ) ., After 1 d ( 24 °C ) or 4 d ( 12 °C ) , 12 juvenile nematodes were transferred at room temperature to individual dishes ( 3 cm diameter ) ., Dishes were randomised and put back at the appropriate temperature ., After 38 h ( 24 °C ) or 145 h ( 12 °C ) dishes were scanned at room temperature at regular intervals ( 1 . 5 h for 24 °C and 4 h for 12 °C ) for the presence of eggs ., If one or more eggs were observed , time and number of eggs were noted and the dish was put at −20 °C to prevent further development; a pilot study ( unpublished data ) showed that freezing did not affect body size ., Dishes were defrosted and nematodes were transferred to new dishes with NGM-agar ., Digital pictures were taken with a CoolSnap camera ( Roper Scientific Photometrics , http://www . photomet . com ) ., Each nematode was measured automatically with Image Pro Express 4 . 0 ( Media Cybernetics , http://www . mediacy . com ) ., Using a measurement ocular we calibrated 10 . 000 pixels3 as 753 . 516 μm3 ., We assumed a rod-like shape of a worm where volume Vszm = π· ( D/2 ) 2·L = ( 1/4 ) ·π·A2/L where D is diameter , L is length , and A = L·D ., Because perimeter P = 2L + 2D ~ 2L we get:, Area ( A , pixels2 ) and perimeter ( P , pixels ) of each worm were measured digitally ., In subsequent analyses Vszm was used as input value for body size 26 ., We studied the TRB , which is a plot of body size versus temperature , and used the slope of the reaction norm as a mapping trait ., For mutant phenotyping the following strains were used for body-size measurements at 24 °C and 12 °C: wild-type Bristol N2 and CB4856 isolate , tra-3 ( e2333 ) , tra-3 ( e1107 ) /dpy-4 ( e116 ) IV , and tra-3 ( e1107 ) sup-24 ( st354 ) IV ., The tra-3 ( e1107 ) /dpy-4 ( e116 ) IV heterozygotes segregate dpy-4 homozygotes , heterozygotes , and tra-3 ( e1107 ) homozygote hermaphrodites , which due to maternal effects are phenotypically wild type and segregate pseudomales 51 ., We measured body size in these homozygote pseudomales , as well as the homozygote and heterozygote hermaphrodites ., Body size was larger only in the hermaphrodites at 12 °C ., Body size in the pseudomales was measured after the characteristic male tail 52 was completely formed ., Experiments were performed on agar dishes ( 3 cm diameter ) as described above ., Samples were not frozen , but body size was measured directly when worms started laying eggs ., Crosses with the mutants and NIL WN17–9 were conducted by transferring J2 stage worms on small agar dishes ( 3 cm diameter ) with three to five males ., The worms were allowed to mate at 24 °C after which the females were transferred to new plates thus allowing them to lay eggs for 3–4 h ., Mating was considered to be successful if the ratio of males:hermaphrodites was approximately 1:1 in the F1 hybrids ., After this period females were removed and eggs allowed to develop at subsequent high or low temperature ., When reproduction started body size was measured as described above ., TG ( Sigma , http://www . sigmaaldrich . com ) was applied to agar plates dissolved in DMSO ., Different concentrations were added in a volume of 200 μl to petri dishes ( 3 cm diameter ) each containing 2 ml of agar ( end concentration in the agar: 0 . 004 , 0 . 0075 , and 0 . 015 μM ) and seeded with E . coli ., A positive control was included containing 200 μl of DMSO ., After 24 h eggs were transferred to each dish and allowed to hatch ., The size at maturity was recorded as described above ., The number of replicate worms measured for their body size were at 24 °C ( italics ) and 12 °C ( bold ) : tra-3 ( e1107 ) 24 , 16; tra-3 ( e1107 ) 24 °C DMSO 10; tra-3 ( e1107 ) 24 °C TG , 10; tra-3 ( e1107 ) sup-24 ( st354 ) 11 , 13; tra-3 ( e2333 ) 19 , 10; +/+ DMSO control 5; +/+ 0 . 004 μM TG 6; +/+ 0 . 0075 μM TG 6; +/+ 0 . 015 μM TG 6; NIL/+ 11 , 15 16 , 11; e1107/+ 8 , 9 , 8 , 8 , 8 , 9; e1107/NIL 10 , 16 , 10 , 18 , 6 , 4; NIL/NIL 22 , 31 ., Populations of N2 and CB were bleached ( 0 . 5 M NaOH , 1% hypochlorite ) to collect synchronized eggs , which were then inoculated into fresh dishes ., For each wild-type strain , four replicate dishes of synchronized eggs were kept in each of the two temperatures until maturity was reached ., The nematodes were then collected and frozen in liquid nitrogen ., Three independent samples were used for each strain and temperature ., For each sample , individuals were synchronized and RNA was extracted using the Trizol method ., RNA was subsequently purified ( with genomic DNA digestion step ) with the RNeasy Micro kit from Qiagen ( http://www . qiagen . com ) ., RNA concentration and quality were measured with Nano Drop ( http://www . nanodrop . com ) ., From each sample 2 μl of RNA were used to obtain cDNA using Superscript II reverse transcriptase from Invitrogen ( http://www . invitrogen . com ) and oligo d ( t ) primers from Genisphere ( http://www . genisphere . com ) ., cDNA was diluted 20× and used for RT-PCR with iQ Sybr Green Supermix from Bio-Rad in 20 μl reactions ( http://www . bio-rad . com ) ., Standard curves for each sample were generated by serial dilutions of the cDNA to select for primer efficiencies of 90%–110% and correlation coefficients greater than 0 . 99 ., We selected two reference genes ( rps-20 and rpl-3 ) using geNorm on the basis of Vandesompele et al . 53 ., All primers were designed with Beacon Designer avoiding secondary structures and cross homology ., RT-PCR runs were done with MyIQ from Bio-Rad , and expression levels were calculated with the Bio-Rad Gene Expression Macro version 1 . 1 using the selected reference genes for normalization ., Expression levels are presented relative to the lowest expression of the gene ., At least two independent experiments were carried out for each gene ., Transgenic worm strains containing tra-3 ( + ) from the Bristol N2 wild-type strain in the CB background were obtained from the Umeå Worm/Fly Transgenic Facility ( http://www3 . umu . se/utcf/index_eng . html ) ., Standard microinjection methods were used 54 ., A DNA fragment spanning the entire tra-3 locus and containing the endogenous tra-3 promoter was injected at a concentration of 25 μg/ml ., The coinjection marker was pCC 55 , a plasmid containing gfp under the control of the unc-122 promoter , which is active in coelomocytes ., pCC was injected at a concentration of 50 μg/ml ., Body size was measured as described above for five independently derived strains of CB ( gfp ) ( control strains ) and CB ( gfp and tra-3+ ) ., For RIL analysis a randomised block design was used ( three blocks per RIL ) ., Statistical analyses were performed in SAS ., All data were found to be normally distributed according to the Box-Cox method ., Comparison between treatments was tested with one-way ANOVA using PROC MIXED ., In case of crossing experiments , replicate crossings were performed , and the data were analysed with a nested design where each cross was nested within temperature ( crosstemperature ) ., In PROC MIXED we defined cross ( temperature ) as a random factor ., The number of replicates was optimal to obtain the mean to be within the 95% CI ., ANOVA was performed to study the effect of temperature , RIL , block , and interactions on body size ., QTL mapping was used to identify the genomic regions ( Wormbase release WS100 ) controlling various life-history traits ., Composite interval mapping was used to identify responsible QTL because it is statistically a wel

Introduction, Results/Discussion, Materials and Methods

Ectotherms rely for their body heat on surrounding temperatures ., A key question in biology is why most ectotherms mature at a larger size at lower temperatures , a phenomenon known as the temperature–size rule ., Since temperature affects virtually all processes in a living organism , current theories to explain this phenomenon are diverse and complex and assert often from opposing assumptions ., Although widely studied , the molecular genetic control of the temperature–size rule is unknown ., We found that the Caenorhabditis elegans wild-type N2 complied with the temperature–size rule , whereas wild-type CB4856 defied it ., Using a candidate gene approach based on an N2 × CB4856 recombinant inbred panel in combination with mutant analysis , complementation , and transgenic studies , we show that a single nucleotide polymorphism in tra-3 leads to mutation F96L in the encoded calpain-like protease ., This mutation attenuates the ability of CB4856 to grow larger at low temperature ., Homology modelling predicts that F96L reduces TRA-3 activity by destabilizing the DII-A domain ., The data show that size adaptation of ectotherms to temperature changes may be less complex than previously thought because a subtle wild-type polymorphism modulates the temperature responsiveness of body size ., These findings provide a novel step toward the molecular understanding of the temperature–size rule , which has puzzled biologists for decades .

Biologists are fascinated by variation in body size , which is hardly surprising , considering that the range of body sizes spans orders of magnitude from bacteria to blue whales ., Even within species , body sizes can vary dramatically ., This intraspecies variation is intriguing because it suggests strong associations between body size and environment ., Already in 1847 , Bergmann noticed that mammals tend to be larger in colder environments ., More recently similar relationships were found for ectotherms , which rely for their body heat on the temperature of their surroundings , where more than 85% of the species studied grew larger at lower temperatures ., This phenomenon , dubbed the temperature–size rule , has caused a renewed interest to understand how temperature affects body size ., Yet the control of the temperature–size rule remains enigmatic , and the hypotheses proposed have been inconclusive ., In this paper the authors show that a single nucleic acid change in one gene is required for regulation of the temperature–size rule in the nematode C . elegans ., Using protein modelling they also show that this subtle change in DNA decreases the function of the encoded protein ., The data suggest that temperature adaptation can be simple and far less complex than previously thought .

developmental biology, ecology, caenorhabditis, evolutionary biology, genetics and genomics

journal.ppat.1005965

A Viral Immunity Chromosome in the Marine Picoeukaryote, Ostreococcus tauri

Eukaryotic micro-algae of the genus Ostreococcus and related species of the order Mamiellales are globally distributed in the photic zone of the worlds oceans where they contribute to fixation of atmospheric carbon and production of oxygen , besides providing a primary source of nutrition in the food web 1–3 ., Their tiny size ( 1–3 μm ) , simple cells ( one chloroplast , one mitochondrion ) , ease of laboratory culture and extremely small genomes , several of which have been completely sequenced 4–8 , render them attractive as models for marine ecology 9–11 , cell biology 12 and evolution in the green lineage 4 , 13 ., Typical features of the highly streamlined genomes of the Mamiellales include a higher GC content ( 48–64% 7 , 14 , 59% for O . tauri ) than higher plants ( 41% , 15 ) and two unusual “outlier” lower GC chromosomes that can carry higher proportions of transposons and genes predicted to originate from prokaryotes ( see 16 for a recent review ) ., In the type species O . tauri , these two chromosomes are chromosome 2 ( the big outlier chromosome or BOC ) and chromosome 19 ( the small outlier or SOC ) , both of which greatly vary in size between strains isolated from the environment 17 ., In marine environments , unicellular organisms account for the largest biomass , but they are outnumbered by about ten to one by viruses , which can lyse host cells and thereby contribute to biogeochemical cycles by promoting the turnover of plankton populations 18–20 ., While large DNA viruses infecting algae have been known for many years ( see 21 , 22 for reviews ) , including viruses of Micromonas pusilla , a member of the Mamiellales 23 , viruses of Ostreococcus spp ., were discovered more recently ., Complete genomes of many viruses infecting Mamiellales ( genus Prasinovirus ) are now available 24–27 ., Given the size of an Ostreococcus cell ( about 1 μm ) prasinoviruses are huge icosahedral particles , about one eighth of the host cell diameter between faces ( 110 nm ) , carrying a repertoire of about 250 genes ., Resistance to prasinoviruses arises spontaneously in cultured algal lines 28 and their complex host–strain specificity patterns 29 , 30 witness their involvement in the planktonic arms race that assures for rapid turnover and evolution in micro-algal populations ( see 31 for a review ) ., Viruses of algae in general and prasinoviruses in particular play an important role in controlling phytoplankton populations 32–36 , the effectiveness of host defence responses thus determines the fate of algal populations ., Regrowth of eukaryotic microalgae following viral lysis with large dsDNA viruses has been observed in phylogenetically diverse lineages 37 , 38 and such virus-resistant lines can be stable in culture 28 , 38 , 39 ., In the diplont marine bloom-forming haptophyte Emiliania huxleyii viral infection of diploid cells led to the selection of haploid cells which were resistant to infection 40 ., Cell lines of the toxic bloom-forming dinoflagellate Heterocapsa circularisquama became resistant to viral infection by HcRNAV after co-culture with this single-stranded RNA virus ., In this case , resistance appeared to be reversible and a variable proportion of resistant cells in culture harboured viruses and produced particles 41 ., Some of these authors suggest that changes in the host cell surface might occur , or that non-infectious defective particles in culture might occupy available receptor sites on host cells , but no specific molecular mechanism has been demonstrated ., In O . tauri , Thomas et al . 28 observed two kinds of virus-resistant cell lines , either lines which were resistant but chronically infected , producing viruses at a low level in culture ( “resistant producers” referred to as RP ) , or cell lines devoid of viruses and immune to re-infection ( “resistant non-producers” , or RNP ) ., Here we aimed to unravel the molecular mechanisms underlying viral resistance using clonal lines of O . tauri RCC4221 4 and its virus , OtV5 ., We produced numerous independent OtV5-resistant lines to examine host and viral gene expression in detail with RNA-Seq technology and to test whether altered gene expression could be responsible for the resistance phenotypes ., The experimental schema in Fig 1 shows how independent clonal lines of O . tauri resistant to OtV5 were generated from a clonal starting population ., The parent OtV5-susceptible O . tauri was subcultured into 46 parallel independent culture lines ( one set of 30 and another set of 16 ) ., OtV5 was introduced into 38 of the O . tauri lines while eight were maintained as non-infected controls ., All cultures with added OtV5 lysed , appearing clear of visible cells within 3–5 days ., The cultures exposed to OtV5 showed regrowth of cells after approximately one week and , upon re-exposure to OtV5 , were subsequently resistant to lysis ., Periodic re-testing for OtV5 resistance showed viral resistance persisted over the almost two-year duration of the study while the control lines continued to be susceptible ( 8 susceptible control lines were maintained alongside the resistant lines shown in Fig 1 , and at each of the 10 dates shown 8/8 lines remained susceptible to OtV5; at the end of the time course , all of the remaining 36 lines were found to be resistant ) ., Resistant lines were periodically tested for production of infective OtV5 by taking cell-free culture medium from resistant lines , adding it to the susceptible parent and observing lysis ., The majority of resistant lines were RP at the start of the experiment , however , the proportion of RP diminished over time so that RNP lines were more abundant at the end of the study ( Fig 2 ) ., Twenty-three O . tauri transcriptomes were sequenced from the experiment including 19 virus-resistant and four susceptible ., To test if OtV5-resistance was linked to differential gene expression and to identify the genes involved , we compared four resistant to four susceptible transcriptomes from matching RNA library batches ( S1 Table ) ., At this early point in the study all but two lines were RP , so resistant lines used in the comparison were RP , however over the course of the study several lines of evidence indicated RP and RNP were probably due to similar molecular mechanisms ( discussed below ) ., Over 95% ( 7432 of the 7749 ) of O . tauri chromosomal genes analysed were transcribed to some extent ( ≥10 aligned reads in at least one sample ) in both resistant and susceptible O . tauri lines suggesting that the majority of genes were expressed at the time of sampling ., A total of 170 O . tauri chromosomal genes were significantly differentially transcribed in resistant lines of which 103 were up-regulated and 67 down-regulated ( Fig 3 , and see S2 and S4 Tables for full gene descriptions ) , which represents only 2% of all expressed O . tauri genes ., Most strikingly , 49 differentially transcribed genes , representing almost a third of differentially transcribed genes were concentrated on chromosome 19 , the SOC ( Fig 3A ) ., These genes on chromosome 19 also had overall larger log2 fold change values indicating larger changes in expression levels on genes from this chromosome than the others ( Fig 3B ) ., Predicted glycosyltransferases ( GTs ) appear to be enriched on chromosome 19 compared to the other chromosomes , the majority of which were differentially transcribed in OtV5-resistant lines ( Fig 3C ) ., The only other chromosome encoding GTs regulated in virus resistant lines was chromosome 2 , the BOC , where the GTs were down-regulated ., Annotated genes on chromosome 19 are known to belong to few functional categories 5 relating to surface membrane proteins , the building of glycoconjugates , as well as methyltransferases ( MTs ) ( Fig 4 , S2 Table ) ., The majority ( 16 of 22 ) of differentially transcribed carbohydrate transport and metabolism genes ( Fig 3D ) were located on chromosome 19 ., Furthermore , almost half of the differentially transcribed genes of unknown function ( 29 of 54 ) ( Fig 3D ) were also encoded by chromosome 19 ., Thus , genes involved in virus immunity in O . tauri are preferentially encoded by the SOC; these genes are strongly associated with carbohydrate modification and metabolism and their specific regulation was associated with viral resistance ., Other functional categories that were significantly differentially transcribed were related to translation , transcriptional regulation ( chromatin remodelling , RNA modification , transcription factors ) , protein modification and turnover , amino acid transport and modification , and other transporters ( Fig 3C ) ., Amino acid biosynthesis genes involved in four different pathways were down-regulated in resistant lines , implicating a decrease in a broad range of amino acids and their downstream metabolites during viral immunity ( S4 Table ) ., Ribosomal subunits were also under-expressed suggesting a slowing of growth-related processes ., Genes involved in expression regulation were likely involved in maintenance of the virus-resistant state ., In particular , histone modification genes were all over-transcribed in the resistant lines pointing to a key role for chromatin restructuring in resistance ., Proteasome-mediated protein degradation ( ubiquitination ) as well as post-translational modification ( glutathione-associated and chaperones ) were up-regulated and may be part of the viral immunity response or translational-level regulation ., Transporters whose substrates were related to undefined small , generally inorganic molecules had mixed regulation with a few cases of transporters with related functions being regulated in opposing directions ., For example , a transmembrane phosphate transporter ( ostta06g00210 ) over-expressed while another calcium-dependent phosphate transporter ( ostta17g00940 ) was under expressed ( S4 Table ) ., Altered transcription of transporters may modulate the available substrate pool required by OtV5 curbing viral replication ., However , these genes show lower significance levels of differential expression and may be involved in processes that are not directly relevant for resistance ., Re-sequencing of the O . tauri parent line with single molecule PacBio technology was able to resolve a large inverted repeat region ( LIRR ) on chromosome 19 totalling ~180 kb ( Fig 4 ) that was predicted to exist from previous work 4 ., The assembled PacBio contig was 311 , 428 bp , in agreement with its expected size from gel mobility ( Fig 5 ) ., Notably , all over-transcribed genes in the resistant lines from the SOC were located in this region ( Fig 4 ) ., Differentially transcribed genes on other chromosomes ( S4 Table ) were in some cases adjacent to each other , but were not grouped together as on chromosome 19 ., Remarkably , transcription of blocks of genes in the LIRR was almost completely suppressed in controls indicating close to half the SOC was silenced in the OtV5-susceptible wild-type state ., By contrast , in resistant lines , these genes were highly transcribed , having among the highest mean fragment counts of the differentially transcribed genes ( S2 Table ) ., The co-transcription of genes on the LIRR suggests they were under the control of the same regulatory factors ., Re-sequencing was also able to resolve a tandem repeat region ( TRR ) comprised of ~2 , 255 bp monomeric repeats sharing 97–99% identity ( S1 Fig ) , which was under-transcribed in resistant lines but still showed some transcription with high sample–sample variance ( Fig 4 ) ., Although small genes of unknown function ( ostta19g00035 and ostta19g00240 ) occurred in the TRR , the sequences outside of these genes appeared transcribed , suggesting a functional role ., The repeats were interrupted at two loci by unique sequences; ( 1 ) a 2 , 232 bp putative terminal repeat retrotransposon in miniature ( TRIM ) possessing long terminal repeats ( LTRs ) and ( 2 ) a 3 , 354 bp sequence of unclear origin ( S1 Fig ) ., The putative TRIM is likely derived from the complete 5 , 537 bp LTR retrotransposon , retrostreo2 on chromosome 8 , with which it shares 99% identity in the LTRs ., Extensive deletions in the intervening gene sequences suggests the putative TRIM is non-autonomous and is an example of LTR-transposon miniaturisation in O . tauri , as well as transposition between chromosomes ., Tandem repeats interrupted by transposons are hallmarks of centromeric repeats 42 , 43 ., However , the repeat sequence was not found on other chromosomes , as would be expected in centromere recognition sites , leaving the functional significance of the TRR unclear ., We conducted an exploratory analysis of all sequenced transcriptomes ( S1 Table ) to see if the chromosome 19 genes detected in the differential expression analysis were consistently regulated the same way in independent resistant lines that were not part of the comparative analysis ., Hierarchical clustering of the long inverted repeat region ( LIRR ) gene transcription profiles grouped all susceptible samples separately from the resistant samples ( S2A Fig ) ., In particular , two blocks of genes ( ostta19g00070–120 and ostta19g000560–610 ) were transcribed across all independent resistant lines , albeit with variation in transcription levels , but not in the susceptible controls , supporting the LIRR having a shared regulatory mechanism activated by OtV5 infection ., By contrast , down-regulated genes were not consistently under-transcribed in all resistant lines ( S2B Fig ) ., This indicates over-expression of the LIRR genes was a strong determiner of resistance while under-expression of chromosome 19 genes , including the TRR , was not consistently correlated with resistance ., Several GT genes from chromosome 19 induced in the virus-resistant state were of putative foreign origin ( S2 Table ) and appeared to be clustered with genes encoding functionalities related to sugar metabolism and modification ( Fig 4 , left gene map ) ., These genes include a rhamnan synthesis F/Wbx GT ( ostta19g00070 ) , family 92 GT ( ostta19g00600 ) and the co-regulated sugar modification enzymes ( ostta19g00110 , NAD-dependent epimerase/dehydratase; ostta19g00610 , CMP-N-acetylneuraminic acid hydrolase ) ., Ostta19g00070 comprises merged rhamnan synthesis F and Wbx GT domains ., The former is allied with bacterial rhamnose-glucose polysaccharide F ( RgpF ) , which in Streptococcus transfers rhamnose to the nascent rhamnan backbone , which is incorporated into surface lipopolysaccharide 44 ., RgpF domain in O . tauri is currently the only occurrence in eukaryotes , the species distribution being otherwise restricted to bacteria ( Pfam: PF05045 ) ., Similarly , Wbx GT domain is found in bacterial gene clusters involved in synthesis of O-antigen , which in Shigella comprises repeated monomers of L-rhamnose , D-galacturonic acid and N-acetylgalactosamine residues 45 , 46 ., One putative gene cluster ( ostta19g00110–140 ) comprises a predicted NAD-dependent epimerase/dehydratase , glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase , integral membrane galactosyltransferase and triose-phosphate transporter ., Strikingly , the 3-beta-galactosyltransferase was associated with sequences from metazoans , while the adjacent genes appear to have homologues in the green lineage ., A second putative gene cluster was a CAZy ( carbohydrate-active enzymes database ) family 92 GT containing a triose phosphate transporter domain ( ostta19g00600 ) adjacent to CMP-N-acetylneuraminic acid hydrolase ( ostta19g00610 ) of metazoan origin ., These two genes were putatively involved in the activation of sugars for the synthesis of cell surface sialic acid ., In proximity to these clusters of sugar modifcation genes was a predicted FkbM methyltransferase ( ostta19g00560 ) , which suggests it has a role in glycan methylation ., Two of the up-regulated GTs ( ostta19g00130 and ostta19g00630 ) , were predicted to be membrane-associated and could be part of the Golgi mannosyltransferase complex ., Two different GTs and a triose-phosphate transporter on chromosome 19 appear to be recently generated paralogous copies because they share up to 97% amino acid identity ( S3 Fig ) ., The paralogues curiously occur in distinct genetic contexts ( Fig 4 ) ., For example , the previously mentioned 3-beta-galactosyltransferase ( ostta19g00120 ) is adjacent to sugar modification genes and is over-transcribed in resistant lines while the putative paralogue ( ostta19g00320 ) is adjacent to a conserved eukaryotic algae hypothetical protein and was under-transcribed ( Fig 4 , GT a ) ., In this case , it suggests the same , or similar , base sugar substrate is utilised in both conditions , but alternatively modified in virus resistant lines ., How these GTs and triose-phosphate transporters have been duplicated and shifted in this modular fashion is unclear , but the transposon-related genes on chromosome 19 are candidates for mediating these transfers ., The only two GTs that were differentially regulated in resistant lines that were not located on the SOC , were located on chromosome 2 ., These genes were down-regulated and included glycosyltransferase AER61 ( ostta02g00040 ) and a membrane glycosyltransferase ( ostta02g02060 ) ., However , other carbohydrate modification and metabolism genes located on chromosome 2 that were up-regulated in resistant lines included a somatomedin B domain-containing protein ( ostta02g03940 ) and carbohydrate glycoside hydrolase ( ostta02g04570 ) ., The former is involved in binding polysaccharides and the latter in the lysis of O-glycosidic bonds ., This implicates a role also for the BOC in carbohydrate modification during viral resistance ., As almost all lines were RP at the time of RNA sequencing , we explored the transcriptomes in sequenced samples for evidence of OtV5 transcription ( S1 Table ) ., In susceptible controls , a low number of read counts were assigned to few OtV5 genes ( S4 Fig ) ; in particular , OtV5_154c and OtV5_159 , which correspond to highly transcribed viral specific genes ( S3 Table ) ., As susceptible cells had not been exposed to OtV5 , apparent transcription of viral genes were taken to be artefacts probably arising from mis-assigned sequencing sample indexes to highly expressed transcripts within the shared sequencing flow cell lane ( all samples were multiplexed together ) ., BLAST searches at neither nucleotide levels nor protein levels revealed similarities that would intimate horizontal transfer of these genes between host and virus ., As expected , lines that were RNP had no OtV5 transcription and clustering with susceptible controls in their OtV5 transcription profiles ( S4 Fig ) ., The RP lines fell into three clusters corresponding to relatively high ( five samples ) , moderate ( five samples ) and negligible OtV5 transcription ( seven samples near-identical to controls ) showing that for most RP lines OtV5 transcripts were nearly undetectable ., As some OtV5 genes appeared to be highly transcribed in certain RP lines , we sought to estimate the amount of virus transcription relative to that of the host ., The ratio of the mean transcript counts of the most expressed viral gene ( OtV5_154c: 5 , 878 ) over that of one of the most highly expressed host genes , the ribulose bisphosphate carboxylase small subunit ( rbcS , ostta18g01880: 49 , 249 ) , gives 0 . 12 ., As a comparison , the transcript abundance ratio of the corresponding genes during Paramecium bursaria chlorella Virus 1 ( PBCV-1 ) infection of Chlorella was 37 , far exceeding host transcription levels 47 ., Since relative OtV5 transcription was lower than that of the host , even in RP lines with relatively higher OtV5 transcripts levels , this suggests viral activity was certainly much lower than expected during infection of susceptible O . tauri ., Nonetheless , the OtV5 genes that were transcribed to a significant level ( mean normalised count > 9 ) covered over 40% of all OtV5 genes ., Expressed OtV5 genes were distributed along the entire viral genome with the expression profile varying for each sample ( S5 Fig ) ., Although we could not compare the OtV5 expression in RP with control cells in a normal lytic cycle to confirm this , the OtV5 transcription profiles did not indicate specific genes linked to chronic infection , nor a particular stage in viral replication ., Crucially , transcripts of capsid proteins were detected ( S3 Table ) consistent with OtV5 forming virions ., This corroborates with results of the virus production assay , which detected infective lytic virions in the medium of RP lines ( Fig 2 ) ., Apart from virion structure , transcribed OtV5 genes were associated with DNA replication , transcription , amino acid synthesis and carbohydrate modification and metabolism ( S3 Table ) ., Expressed OtV5 carbohydrate metabolism genes included two enzymes involved in the biosynthesis of nucleotide sugars , OtV5_011 , a predicted GDP-D-mannose 4 , 6-dehydratase , and OtV5_042 , a putative dTDP-D-glucose 4 , 6-dehydratase ., The homologue of this GDP-D-mannose 4 , 6-dehydratase in PBCV-1 has a dual functionality acting in both the synthesis of GDP-D-rhamnose and GDP-L-fucose , both of which are monosaccharides present in the capsid glycan but are rare in the host 48–50 ., Adjacent to the GDP-D-mannose 4 , 6-dehydratase is a group 1 GT ( OtV5_012c ) that was also expressed ., The spatial proximity of these genes suggests the glycosyldonor of the group 1 GT is the nucleotide sugar produced by the GDP-D-mannose 4 , 6-dehydratase ., Three additional OtV5 GTs were transcribed including a predicted membrane-localised group 34 GT ( OtV5_033 ) , a group 2 diphosphosugar GT ( OtV5_035 ) and a GT with no assigned CAZy group ( OtV5_160 ) ., All OtV5 encoded GTs had their closest homologues in prasinoviruses indicating they have conserved virus-specific functions ., Given the massive transcriptional changes in chromosome 19 and the known plasticity in outlier chromosome size between O . tauri strains 17 , we examined the karyotypes of all experimental lines by pulsed-field gel electrophoresis ( PFGE ) ., Resistant lines showed karyotype changes , most notably as a shift in the size of chromosome 19 ( 34 of 36 resistant lines tested ) while no change was evident in the susceptible controls ( Fig 5 ) ., The most common changes were an increase of ~20–490 kb ( 17 lines ) or a decrease of ~40–140 kb ( 13 lines ) in the size of chromosome 19 ., This was likely due in the former case to duplications within chromosome 19 and in the latter , to deletions within , or fission of , chromosome 19 ., Several resistant lines ( R1 , R22 , R29 and R30 ) appeared to have lost chromosome 19 or its location was ambiguous ., In R9a and R2 , there were possible translocations of regions of chromosome 19 to chromosomes 6 and 8 ( respectively ) or an increase in the size of chromosome 19 with concurrent insertions-deletions in other chromosomes ., Indeed , variations in the size of chromosomes other than chromosome 19 were noted in R11a , R3 , R5 , R22 , R23 and R30 ., More complex changes were also evident , most notably in R30 where bands apparently larger than 1 . 1 Mb are present that hybridize with the chromosome 19 probe ., Interestingly , seven resistant lines ( R9a , R13a , R1 , R2 , R12 R18 and R26 ) showed densely hybridizing material in the wells , almost always coinciding with a large increase in the size of chromosome 19 ( >430 kb ) ., This could correspond to large circular forms that cannot migrate in the PFGE 51 , suggesting the extensive changes to chromosome 19 could occur via circular intermediates ., However , the presence of circular DNA in the nucleus such as episomes has neither been previously detected nor specifically investigated in O . tauri ., To investigate the chronic production of OtV5 , electron microscopy was performed on two RP lines ., This showed a minority of cells ( <0 . 5% ) was visibly infected and/or in the middle of lysis ( Fig 6 ) demonstrating viral reproduction proceeded by a typical lytic cycle ., Nonetheless , a population crash of the resistant cultures was never observed , with the only loss of lines occurred following antibiotic treatment , so the frequency of lysis was low ., Since all lines were re-cloned upon acquiring resistance ( Fig 1 ) , the minority of susceptible cells present in RP lines were presumably resistant cells that had switched to a susceptible state ., Production of the OtV5 genome was also observed by PFGE as a ~200 kb band ( Fig 5 ) in 15 of the 19 lines that were RP at that time point ., In the four RP lines where the band was not apparent , viral genomes were likely present below the detection limit of DNA staining ., The intensities of the viral bands were both greater and less than that of the host chromosomes suggesting , ( 1 ) a large variability in amount of viral production between RP lines and ( 2 ) , given that the number of genomes in the sample is proportional to band intensity , the number of copies of the viral genome was not directly proportional to that of the host ., The fact that the OtV5 genome copy number could be much lower than that of the host implies not every cell in the host population was carrying the virus genome ., Higher OtV5 band intensities could correspond to viral genomes in virions as well as multiple viral genomes generated during replication in the cell ., Outlier chromosomes are found in all Mamiellales sequenced to date , and all SOCs encode genes with a similar set of predicted functions 7 , although these genes have very little detectable sequence homology compared to the non-outlier chromosomes ., Only 11% of genes on the SOCs of O . lucimarinus and O . tauri are orthologues , whereas the average proportion of orthologous genes is 86% on standard chromosomes 52 ., We re-analysed the functional genes in all available Mamiellales genomes observing a consistent over-representation of genes involved in carbohydrate transport , synthesis and modification , particularly GTs ( Table 1 ) ., Prasinoviruses are known to infect all clades of Ostreococcus , Micromonas , and Bathycoccus so far tested 29 ., Furthermore , prasinovirus resistance has also been generated in Bathycoccus prasinos against BpV2 ( Bathycoccus prasinos Virus, 2 ) and Micromonas pusilla against MpV1 ( Micromonas pusilla Virus1 ) in culture 28 ., Given that differentially expressed genes on the SOC in O . tauri was strongly linked to a switch to a virus-immune state , we hypothesize the SOC has a similar role in virus resistance in these other genera of the Mamiellales ., Initially , we designed the experimental strategy to test whether resistant cells may arise spontaneously by mutation , and we planned to look for single nucleotide changes linked to resistance ., However , the first visible signs of re-growth of resistant cells appeared about a week after inoculation in all of the cultures ., Since cultures do not become visibly green until they reach at least 106 cells . ml-1 , and assuming the cells have an optimum division rate under these conditions ( 1 . 4 divisions . day-1 in this growth chamber , M . Krasovec , personal communication ) , we find that about 1 in 1 , 000 cells may have become resistant at the time of or just after inoculation with the virus ., This , however , may be an underestimation , since after a shock ( such as after sub-culturing ) , there is usually a lag before re-growth ., This frequency far exceeds the expected spontaneous mutation rate in O . tauri since comparison of the re-sequenced O . tauri genome between 2001 and 2009 revealed the fixation of eight single nucleotide substitutions and two deletions during the approximately 6 , 000 generations in the lab 4 ., We thus preferred an alternative hypothesis , that resistance is induced by the biotic challenge of virus infection , for example by epigenetic modifications affecting gene expression patterns ., Since the original culture was clonal , it seemed unlikely that a proportion of host cells was resistant at the moment of inoculation; although we cannot rule out that a reversible regulatory switch to a resistant state occurred in a small proportion of cells analogous to the minority of cells in RP lines that was lysed by OtV5 ., However , after a period of selection , resistance in the majority of cells appears to be stable since none of the resistant lines were lysed after re-infection over the course of the experiment ., In several loci , up-regulated genes on the LIRR related to carbohydrate metabolism and modification appeared to be spatially grouped ., We speculate genes that are clustered together act on the same carbohydrate synthesis and glycosylation pathway ., Notably , genes uniquely expressed in virus-resistant lines might dramatically alter the saccharide composition and glycosylation state of the cell ., Several of the highly expressed GTs are associated with synthesis of surface glycans ., Surface glycans are known to be important for host–virus interactions , where they mediate initial binding and recognition events of both host cells and pathogens at the cell surfaces 53 ., In particular , rhamnan is a polymer important for bacterial host–virus interactions 54 , 55 , and sulphated derivatives of rhamnans are known to have anti-viral activity in mammals 56 ., The metazoan-derived CMP N-acetylneuraminic acid hydroxylase ( ostta19g00610 ) , in humans makes the influenza A virus receptor , N-acetylneuraminic acid , whose decorations affect receptor specificity 57 ., Likewise , glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase ( ostta19g00120 ) , known to function in the synthesis of extended mucin type O-linked glycans , has been associated with certain immune mediated diseases in humans 58 ., Thus , expression of these genes may affect resistance through masking , or altering the O . tauri surface receptor used by OtV5 to inhibit viral adsorption ., However , Thomas et al . , 28 observed no statistical difference in OtV5 adsorption to RP , RNP and susceptible cells suggesting defective adsorption was not the mechanism for viral resistance ., This may be due in part the very low proportion of O . tauri cells ‘competent’ to OtV5 adsorption at a given time point; even after inoculating susceptible O . tauri with high titres of OtV5 , at most 20% of cells had visibly adsorbed virions despite the fact the majority of cells subsequently lyse 24 ., One possible explanation is that the OtV5 receptor is only available during a defined point in host cell cycle , and as O . tauri cells are not perfectly synchronised in culture 59 , adsorption is staggered over time making differences difficult to detect by a standard adsorption assay ., Alternatively , the over-expression of these specific host GTs may curtail OtV5 replication at another stage of the replication cycle , one possible candidate being virion assembly ., In the model Chlorovirus , PBCV-1 , the major capsid protein is glycosylated at least six sites and the glycans include rare monosaccharides whose biosynthesis genes are virus-encoded ( see Piacente et al . , 60 and references therein ) ., As OtV5 similarly encodes numerous GTs and homologues to the PBCV-1 sugar metabolism enzymes , many of which are conserved in prasinoviruses ( S3 Table ) , glycoconjugates are likely to be similarly essential to virion structure ., Viral resistance may arise by the host switching carbohydrate metabolic pathways , thereby perturbing proper virion formation ., This could be mediated , for instance , by feedback inhibition of viral sugar biosynthesis pathways by host generated products , directly altering the viral glycosyldonors/acceptors or simply altering the substrate pool ., For example , virally produced GDP-D-rhamnose may be depleted by the host by redirecting it into extracellular rhamnan ., Much remains unknown about the glycobiology of both prasinoviruses and Mamiellales and studies of their sugar composition , glycan structure and functional characterisation of GTs will help to pinpoint how the host–virus interaction is suppressed during resistance .,

Introduction, Results, Discussion, Materials and Methods

Micro-algae of the genus Ostreococcus and related species of the order Mamiellales are globally distributed in the photic zone of worlds oceans where they contribute to fixation of atmospheric carbon and production of oxygen , besides providing a primary source of nutrition in the food web ., Their tiny size , simple cells , ease of culture , compact genomes and susceptibility to the most abundant large DNA viruses in the sea render them attractive as models for integrative marine biology ., In culture , spontaneous resistance to viruses occurs frequently ., Here , we show that virus-producing resistant cell lines arise in many independent cell lines during lytic infections , but over two years , more and more of these lines stop producing viruses ., We observed sweeping over-expression of all genes in more than half of chromosome 19 in resistant lines , and karyotypic analyses showed physical rearrangements of this chromosome ., Chromosome 19 has an unusual genetic structure whose equivalent is found in all of the sequenced genomes in this ecologically important group of green algae .

We propose that chromosome 19 of O . tauri is specialized in defence against viral attack , a constant threat for all planktonic life , and that the most likely cause of resistance is the over-expression of numerous predicted glycosyltransferase genes ., O . tauri thus provides an amenable model for molecular analysis of genome evolution under environmental stress and for investigating glycan-mediated host-virus interactions , such as those seen in herpes , influenza , HIV , PBCV and mimivirus .

chromosome 19, genome evolution, gene regulation, microbiology, dna transcription, microbial genetics, microbial genomics, viral genomics, chromosome biology, gene expression, molecular evolution, chromosome pairs, viral genetics, cell biology, virology, genetics, viral gene expression, biology and life sciences, genomics, evolutionary biology, computational biology, chromosomes

journal.pcbi.1006950

The relative contribution of color and material in object selection

In daily life , we rely on vision to select objects for a variety of goal-directed actions ., For example , when we crave tomatoes , we use color to decide which tomato on the vine is the ripest ( Fig 1A ) ; when we sip coffee , we use glossiness to judge whether a cup is made of porcelain or paper , which in turn affects how we handle it ( Fig 1B ) ., Indeed , we continually use visual information to effortlessly and confidently judge object characteristics ., Instances in which vision misleads us are sufficiently rare to be memorable , as in the case of a deflated basketball sculpture made of glass ( Fig 1C ) ., Extracting information about object properties from the image formed on the retina by light reflected from objects is a challenging computational problem ., This is because the process of image formation entangles information about the intrinsic properties of objects ( such as color or material ) with information about the conditions under which they are viewed ., For example , the retinal image is affected by changes in the illumination , the objects’ position and pose , and the viewpoint of the observer ., Understanding the perceptual computations that transform the retinal image into stable representations of objects and their properties is a longstanding goal of vision science ., A large literature has employed a divide and conquer strategy to investigate the perception of object properties: different object attributes ( color , texture , material , shape , etc . ) have each been studied within their own subfields ., This approach has leveraged well-controlled laboratory stimuli and relatively simple psychophysical tasks to build a quantitative understanding of how information is transduced and represented early in the visual pathways 1 , 2 ., In addition , careful case studies have provided insight into how stable perception of object properties may be achieved when the experimental conditions are relatively simple and well-specified 3–7 ., Our work builds on the foundations provided by this approach and aims to extend the study of object perception in two critical ways ., First , we want to move beyond highly-simplified laboratory stimuli and tasks and devise paradigms in which object perception is probed using both naturalistic stimuli and naturalistic tasks ., Second , to explain real-life object representations , we need to describe how perceptual judgments along multiple dimensions ( color , material , shape , size , texture , etc ) combine and interact ., For example , even though color provides an important cue for selecting the ripest tomato ( Fig 1A ) , other characteristics , such as tomato size , shape , gloss , and surface texture also provide useful information that can guide selection ., Our experimental paradigm employs naturalistic stimuli in combination with a two-alternative forced-choice object selection task ., This task captures a core aspect of how vision is used in real life , where it guides object selection in service of specific goals e . g . , selecting nutritious and avoiding spoiled food , 8 ., We have previously shown how a version of the elemental selection task can be embedded within more complex and naturalistic tasks to probe color perception 9 , 10 ., Here we elaborate the selection task to measure the underlying perceptual representations of both object color and material ( specifically , glossiness ) and quantify how these two perceptual dimensions combine in object selection ., Note that we use the term material to refer to the physical glossiness , which is a function of the geometric reflectance properties of object surfaces ., Similarly , we use the term color to refer to the diffuse surface spectral reflectance of objects ., When it is not clear from context , we will explicitly distinguish the perceptual correlates of these physical properties ( e . g . , perceived material and perceived color ) ., The object selection task is illustrated by Fig 2 ., On each trial , observers viewed three blob-shaped objects—the target and two tests—and selected the test that was more similar to the target ., Across trials , the color and glossiness of the target object was fixed ., The tests were identical to the target in their shape , size and pose , but their color and material varied from trial to trial ., Test color could vary from blue to green , across 7 different levels; test material could vary from matte to shiny , also across 7 levels ( Fig 3 ) ., For each pair of test objects , we measured the probability that each member of the pair was selected ., We report three primary results ., The first is theoretical ., To understand the selection data , we need an observer model that translates the raw data into an interpretable form ., An important advance of the work we present here is the development of such a model ., The model describes the data in terms of how the stimuli are positioned along underlying perceptual dimensions and how distances along these dimensions are combined to guide object selection ., Our second result is experimental ., We show that there are large individual differences in the degree to which observers rely on object color relative to object material in selection ., Some observers base their selections almost entirely on color , some weight color and material nearly equally , and others rely almost entirely on material ., Third , a fine-grained analysis of our data , in parallel with model comparisons , clarifies limits on how precisely selection data may be leveraged to simultaneously reason about perceptual representations and color-material trade-off ., These limits , which we make explicit , are important to recognize as we and others move towards understanding the multidimensional nature of object perception ., Below , we present each of these results in detail ., Our observer model builds directly on our recent work on color selection 8–10 and incorporates concepts from multidimensional scaling 11 , 12 , the theory of signal detection 13 , and maximum likelihood difference scaling 14 , 15 ., As in multidimensional scaling , our model assumes that each stimulus is represented in a subjective perceptual space where , in our case , the dimensions are color ( C ) and material ( M ) ., Rather than using a fixed location to represent each stimulus , we incorporate the idea that perception is noisy e . g . , 13 , 16 and model the representation of each stimulus as a bivariate Gaussian distribution ., The mean of each Gaussian locates the corresponding stimulus in the perceptual space , while the covariance specifies the precision of the representation ., The model assumes that on each trial of the experiment , the actual representation of each stimulus ( target and two tests ) is a draw from the corresponding distribution and that the observer chooses the test stimulus whose representation on that trial is closest to that of the target ., The probability that one test is chosen over another depends on the mean positions of their underlying representations , the magnitude of the perceptual noise , and a color-material weight ., This weight describes how differences along the two perceptual dimensions ( color and material ) are integrated when the observers select objects based on similarity ., In the model , we define the origin of the perceptual space by setting the position of the target to zero on each dimension ., Note that the target remains constant across all trials ( and therefore the origin is fixed ) ., Similarly , we define the scale of the perceptual dimensions by setting the variance of the perceptual noise to one for each dimension ., These conventions do not affect the models ability to account for the data ., The observer’s performance is then described in terms of two key sets of parameters: ( 1 ) parameters that describe physical-to-perceptual mapping i . e . , the mean position of each stimulus in the color-material perceptual space and ( 2 ) the color-material weight , which we denote as w ., The computation of perceptual distances between the target and each test occurs only after distances on the color dimension have been scaled by w and distances along the material dimension have been scaled by 1-w ., The color-material weight thus characterizes the relative importance of object color relative to object material in selection ., The model does include some substantive assumptions ., First , we assume that the perceptual representation of our stimuli is two-dimensional ., Second , we assume that positions along the color and material dimension are independent ., That is , varying the position of a stimulus on the material dimension does not affect its position on the color dimension and varying the position of a stimulus on the color dimension does not affect its position on the material dimension ., Third , we assume that the perceptual noise along the two underlying dimensions is independent ., Fourth , we assume that the noise is additive and independent of the stimulus level ., We return to consider the implications of these assumptions in Discussion ., We considered multiple variants of the perceptual model ., These differed in two ways ., First , we considered two different distance metrics ( Euclidean and City-Block ) for computing overall test-to-target distances , based on the weighted distances along each underlying dimension ., There is large literature in perception and cognition that discusses whether Euclidean or City-Block metric best describes perceived similarity 17–19 ., We did not have an a prori reason to favor one type of metric vs . another and we chose to compare them empirically ., Second , we considered four different ways of mapping nominal stimulus positions ( labeled as -3 to +3 for each dimension , Fig, 3 ) to the corresponding mean perceptual positions ., In the Full model variant , each non-zero nominal label was mapped onto its own mean perceptual position , so that 12 parameters were needed to describe to the mapping ( 6 for each dimension ) ., In the Cubic , Quadratic , and Linear model variants , the mean perceptual positions were obtained from cubic , quadratic and linear functions of the nominal labels ( 6 , 4 and 2 parameters respectively ) that pass through the origin ( target coordinate of 0 , 0 ) ., Thus 8 model variants were considered ( 2 metrics crossed with four positional-mapping variants ) ., Additional details about the model implementation and how it was fit to the data are provided in Methods together with a formal expression of the model ., Our experimental design used Quest+ , an adaptive trial selection procedure 20 , together with the Euclidean/Cubic variant of our model ., Given the parametric model , Quest+ selects for each trial the pair of test stimuli ( 7 levels per dimension , 49 possible stimuli , 1176 possible test stimulus pairs ) that is predicted to yield the most information about the model parameters , given the selection data collected up to that point ., The use of an adaptive method was critical for making the experiment feasible , as we estimate it would have taken ~40 hours per observer to measure the selection probabilities for all possible test pairs ( ~20 trials each for 1176 possible test pairs ) ., Development of the model and experimental procedures were guided by our findings in a preliminary experiment that used a subset of possible trial types ., This experiment is described in a conference proceedings paper 21 , also reviewed in 10 ., For each observer , we used a preregistered model selection procedure based on cross-validated fit error to find which of the 8 model variants best accounted for each observer’s selection data ., A detailed description of this procedure is available in Methods ., We then used the best-fitting model variant to infer the positions of the stimuli in the perceptual color-material space and the color-material weight for each observer ., Fig 4 shows the model solution for three of our observers ., These observers differ in their color-material weight ( dca w = 0 . 12; sel , w = 0 . 45; nkh , w = 0 . 85 ) ., Each row shows data for one observer ., The model solution is represented across three panels , which illustrate the recovered parameters , the quality of model fit , and what we refer to as color-material trade-off functions ., Table 1 indicates which model variant was best for each observer ., The left column of Fig 4 shows the inferred stimulus positions ., The target is located at the origin ., The x-axis shows the color dimension: points to the left of the origin indicate stimuli that are greener than the target and points to the right indicate stimuli that are bluer ., The y-axis shows the material dimension: points below the origin indicate stimuli that are glossier than the target , while points above indicate stimuli that are more matte ., Within each dimension , the mapping between nominal stimulus positions and perceptual positions is ordered ( from C-3 on the left to C+3 on the right and from M-3 on the bottom to M+3 on the top ) ., The inferred stimulus positions differ across observers ., In addition , the inferred stimulus spacing along each dimension is not uniform ., This should not be surprising: without extensive preliminary experimentation there is no reliable way to choose stimuli that have uniform perceptual spacing for each observer ., The center column illustrates the quality of the model fit to the data ., For each stimulus pair shown more than once , the measured proportion of trials one test was chosen relative to another is plotted against the corresponding proportion predicted by the best-fitting model ., The diagonal represents the identity line: the closer the points are to the diagonal , the better the agreement between model and data ., The area of each plotted point is proportional to the number of trials run for a given stimulus pair: the larger the data point the more trials were shown ., The model provides a reasonable account of the data , with the large plotted points lying near the diagonal ., The right column shows color-material trade-off functions ., These are the model predictions for trials in which one of the tests is a color match and the other test is a material match ., We use the term color match to refer to tests that have the same color as the target but differ in material , and the term material match to refer to tests that have the same material as the target but differ in color ., The color-material trade-off functions show the proportion of time a color match is chosen ( y-axis ) , when paired with the material matches ., The color difference of the material match from the target is indicated on the x-axis ., The black line shows the trade-off for a color match that is identical to the target ( zero material difference: M0 ) ., When paired with the material match that is also identical to the target ( zero color difference: C0 ) , predicted selection proportion is at chance ., As the color difference of the material match increases , the predicted probability that the observer chooses the color match increases and approaches 1 ., The red lines show the trade-off for a color match for which the material difference from the target is large ( dashed line: M-3; solid line: M+3 ) ., When paired with the material match that is identical to the target ( C0 ) , the observer is predicted to select the material match ( color match selection proportion near 0 ) ., As the color difference of the material match increases , however , the observer switches to selecting the color match , tolerating the difference in material ., The green and blue lines indicate trade-off functions for intermediate values of color match material difference ( small difference step in blue: M-1 is dashed and M+1 is solid line; medium difference step in green: M-2 is dashed and M+2 , is solid line ) ., These fall between the black and red lines ., The relative steepness of the color-material trade-off functions reflects how readily the observer transitions to preferring the color matches over material matches ., The steepness of the functions also varies across the three observers and is qualitatively consistent with the differences in the inferred color-material weight ., For example , the trade-off functions for the observer nkh , who has a high color-material weight ( tends to make selection based on color ) , indicate very low tolerance for color differences of the material matches before the predicted selections switch to the color matches ., The trade-off functions for observer dca , who has a low color-material weight are flattened , in comparison , indicating a large degree of tolerance for color differences of the material match ., Note , however , that the trade-off functions depend both on the perceived spacing between the stimuli along the color and material dimensions , as well as on the color-material weight ., The relative symmetry in the model solution for most observers ( Fig 4 left and right ) reflects the degree to which test-to-target differences in the ( nominally ) positive and negative directions are perceptually equated ., For a given observer , right-left symmetry of the predicted color-material trade-off functions indicates the degree to which the same-size steps in the positive and negative direction are perceptually equated in the color dimension ( Fig 4 right; also left-right symmetry of positions in Fig 4 left ) ., Similarly , the degree of overlap between the predicted color-material trade-off functions shown in dashed and solid line of the same color indicates the extent to which the same-size steps in the positive and negative direction are perceptually equated in the material dimension ( Fig 4 right; also top-bottom symmetry of positions in Fig 4 left ) ., As we note in Methods , we tried to choose stimulus levels on each dimension that were spaced in a perceptually uniform manner ., This was only approximate , however , and there was no guarantee that the steps would also appear uniform to our observers ., We evaluated the quality of the fit of the color-material trade-off functions to the data by plotting the measured selection proportions for the trials in which a color match test is paired with the material match ., Because the trial selection was determined by the Quest+ procedure , only a subset of such trials was presented and the number of trials per pair varied across observers ., Points are plotted only for pairs shown at least 10 times ., Comparison of plotted points with corresponding prediction lines shows good agreement in most cases ., One of the key goals of our model was to independently describe the underlying stimulus representation in a perceptual color-material space and the color-material trade off ., In other words , we aimed to uniquely determine ( 1 ) the parameters describing the stimulus positions and ( 2 ) the color-material weight ., Two aspects of the results indicate that there are limits on how well this can be accomplished ., First , some of the bootstrapped confidence intervals for the color-material weight are large ( Fig 5 ) ., Second , examination of Fig 4 indicates the possibility of a systematic relationship between stimulus positions and the color-material weight ., For the observer ( dca ) for whom the color-material weight is small , the inferred stimulus positions on the color dimension are expanded relative to those on the material dimension ., For the observer ( nkh ) for whom the color-material weight is large , the opposite relation is seen: positions on the material dimension are expanded relative to those on the color dimension ., To investigate this further , we summarized the relation between the positions on the color and material dimension ( inferred from the best-fitting model ) by first finding for each dimension the slope of the linear function mapping nominal stimulus position labels ( integers between -3 and +3 ) to perceptual positions ., We then computed the ratio of the slope for color to the slope for material ., This color-material slope ratio is large when the positions on the material dimension are compressed relative to the positions on the color dimension ( e . g . , dca ) and small when material is expanded relative to color ( e . g , nkh ) ., Thus , the color-material slope ratio provides an index for relative positional expansion on the two perceptual dimensions , and we can examine how it varies with the color-material weight ., Fig 6 ( top panel ) shows the set of bootstrapped color-material slope ratios against the corresponding color-material weights , with results for each observer shown in a different color ( color-to-observer mapping is shown in the bottom panel ) ., The black open circles show the slope ratio and weight inferred from the best-fitting model applied to the complete data set for each observer ., The figure illustrates that there is a systematic trade-off between the two aspects of the model solution ., Within each observer , the higher the color-material weight , the lower the color-material slope ratio ., The correlation between the two numbers is highly statistically significant for every observer ( Pearson correlation coefficients ranged from -0 . 99 to -0 . 91 , p < 0 . 0001 for all observers ) ., The distribution shown for each observer reflects the measurement uncertainty in determination of the two aspects of the solution ., The confidence intervals shown in Fig 5 represent the central 68% of the x-axis variation for each observer shown in Fig 6 ., Fig 6 demonstrates that stimulus positions and the color-material weight are entangled in the model solution: changes in color-material weight can be compensated for , by adjusting the stimulus positions without a large effect on the quality of the model fit ., In other words , the observer selection pattern can be explained either in terms of a higher weight being placed on color , accompanied by inferred stimulus positions that yield a lower color-material slope ratio , or a lower color-material weight , accompanied by a higher color-material slope ratio ., There are two important features of this parameter trade-off ., First , the degree to which it occurs varies across different observers ., For some observers , the color-material weight is well-determined , despite the parameter trade-off , while for other observers the trade-off limits how well we can determine the color-material weight given our data ., This is summarized by the bootstrapped confidence intervals in Fig 5 , which are obtained from the distribution of the bootstrapped solutions along the x-axis shown in Fig 6 ., Although confidence intervals on the color-material weight are reasonably small for most observers , there are cases for which they are large ( e . g . , observers hmn , cjz and nzf ) , suggesting that for these observers the data do not have sufficient power to determine the color-material weight ., Second , despite the parameter trade-off , Fig 6 also illustrates clear individual differences across observers ., In particular , the distributions of the bootstrapped color-material weight overlap minimally for some observers ( e . g . , green and yellow points vs . pink and lime points ) ., Furthermore , even when the range of color-material weights overlaps , the data for different observers can fall along distinct lines ( e . g . red versus yellow points; pink versus lime points ) : there are individual differences in performance even for observers whose color-material weights are not differentiated by the data ., These differences cannot be ascribed either to differences in perceptual representation or to color-material weight , but rather to some undetermined combination of the two ( differences in stimulus positions if the color-material weight is equated and differences in color-material weight if the color-material slope ratio is equated ) ., A goal for future work is to reduce this ambiguity , as we consider in more detail in Discussion ., Our current modelling clarifies what individual differences can and cannot be forcefully characterized within our experimental framework ., Color , material , texture and shape inform us about objects and guide our interactions with them ., How vision extracts information about individual object properties has been extensively studied ., Little is known , however , about how percepts of different properties combine to form a multidimensional object representation ., Here we describe a paradigm we developed to study the joint perception of two different object properties , color and material ., Our work builds on the literature on cue-combination , which also considers the multidimensional nature of object perception 23–27 ., What distinguishes our approach is that we move beyond threshold measurements to study supra-threshold differences see also 28 , 29 ., On each trial of our object selection task observers viewed objects that vary in color and material ( glossiness ) and made selections based on overall similarity ., We interpret the selection data using a novel observer model ., The model allows us to describe the data in terms of the underlying perceptual stimulus representation and a color-material weight , which quantifies the trade-off between object color and object material in selection ., We find large individual differences in color-material weight across twelve observers: some observers rely predominantly on color when they select objects , others rely predominantly on material , and yet others weight color and material approximately equally ., Although our results show salient individual differences across observers , they also show that for some observers the confidence intervals on the color-material weight are large , encompassing most of the possible 0 to 1 range ., In other words , for three of our observers ( hmn , cjz and nzf ) the color-material weight is underdetermined , given the data ., Currently , we can only speculate about why this occurs ., One possibility is that large confidence intervals emerge because these observers change their selection criteria over the course of the experiment , which may amplify the ambiguity between color-material weight and color-material slope ratio we discuss above ( Fig 6 ) ., Development of our observer model required us to overcome two fundamental challenges ., The first arises because both the underlying perceptual representation of the stimuli and the way information is combined across perceptual dimensions are unknown and thus need to be recovered simultaneously ., Although these two factors are conceptually different , their variation can have a qualitatively similar influence on the observers’ selection behavior ., An important advance of our model is that it allows us to separate the contribution of the two factors ., This separation works sufficiently well to allow us to establish that individual observers employ different color-material weights ., At the same time , our work reveals limits on how precisely the contribution of the two factors can be separated ., Improving the precision this separation represents an important direction for future work ., We return to this point later in the Discussion ., The second challenge arises because as the number of dimensions studied increases and the stimulus range extends to include supra-threshold differences , the set of stimuli that could be presented grows far too rapidly for exhaustive measurement ., This highlights the need for a theoretically-driven stimulus selection method , which would enable estimation of model parameters from a feasible number of psychophysical trials ., To address this challenge , we implemented an adaptive stimulus selection procedure , which incorporated a seven-parameter variant of our model ., The procedure is based on the Quest+ method 20 and selects on each trial the test stimulus pair that is most informative about the underlying model parameters ., The strength of this approach is that it allows us to exploit appropriately complex models of how observers perform our task ., One side-effect of using this efficient procedure , however , is that the power of the data to test the how well the model accounts for performance is reduced ., We handled this by conducting a preliminary experiment as a part of model development 21 ., In this experiment we studied only a subset of stimulus pairs ( color matches paired with material matches ) , using the method of constant stimuli , and showed that our model ( Full positional variant with Euclidean distance metric ) accounts well for the selection data ., As we noted above , our analyses show that in the model solution the recovered stimulus positions and color-material weight are not entirely independent: variation in the color-material weight can be compensated by variation in stimulus positions with minimal effect on the quality of the models fit ., This trade-off in the recovered model parameters emerges because our model explicitly includes perceptual noise along each perceptual dimension ., Counter-intuitively , this means that even when stimuli vary only along a single perceptual dimension ( e . g . color ) , changing the stimulus spacing along that dimension need not have a large effect on observers’ predicted selection probabilities ., Instead , such change in spacing can be compensated by changing the relative weight , which in turn affects contribution of noise in the other ( non-varied ) dimension on the predicted selection probabilities ., Although this compensatory effect is not complete , developing ways to more forcefully identify model parameters remains an important future research direction , as we and others move towards understanding the multidimensional nature of object representations ., Below we outline three possible approaches to address this challenge ., The first approach is to obtain direct measurements of the underlying stimulus representations along each perceptual dimension ., This could be achieved by conducting separate experiments in which observers are explicitly instructed to select objects based on only one aspect of the stimulus at a time ( either color or material ) ., This approach was taken in recent work that focuses on the independence of the perceptual representations of multiple object properties 28–32 ., It relies on the assumption that when instructed to attend to just one stimulus dimension , observers are able to ignore variations along the other dimensions ., If the assumption holds , this approach combined with ours would provide additional power to recover stimulus positions along the attended dimension because it enables fitting the data without introducing the effects of noise along the non-attended dimension ., There is no guarantee , however , that observers can or do strictly follow experimental instructions that direct them to attend only to a single perceptual dimension ., The second approach is to simplify our observer model and assume that the underlying stimulus representation is common across observers , so that any variation in performance is entirely due to the variation in the color-material weight ., While the assumption that all observers perceive the stimuli in the same way may be questioned , it has a long history in the study of perception ., Indeed , this approach is implicit in ( 1 ) efforts to develop standardized perceptual distance metrics and stimulus order systems e . g . , 33 , ( 2 ) studies in which the conclusions about perceptual representations are based on data averaged across multiple observers e . g . , 34 , Fig 2 and ( 3 ) work that employs multidimensional scaling to recover the common perceptual representation across observers together with the parameters that describe different weighting of the underlying dimensions by each observer e . g . , 35 ., Finally , the third approach is to employ a multiplicative rather than an additive noise model ., The precision of perceptual representations is often the highest at the current adaptation point 36–38 ., This observation can be incorporated in the model by assuming that along each perceptual dimension noise scales as a function of test distance

Introduction, Results, Discussion, Methods

Object perception is inherently multidimensional: information about color , material , texture and shape all guide how we interact with objects ., We developed a paradigm that quantifies how two object properties ( color and material ) combine in object selection ., On each experimental trial , observers viewed three blob-shaped objects—the target and two tests—and selected the test that was more similar to the target ., Across trials , the target object was fixed , while the tests varied in color ( across 7 levels ) and material ( also 7 levels , yielding 49 possible stimuli ) ., We used an adaptive trial selection procedure ( Quest+ ) to present , on each trial , the stimulus test pair that is most informative of underlying processes that drive selection ., We present a novel computational model that allows us to describe observers’ selection data in terms of ( 1 ) the underlying perceptual stimulus representation and ( 2 ) a color-material weight , which quantifies the relative importance of color vs . material in selection ., We document large individual differences in the color-material weight across the 12 observers we tested ., Furthermore , our analyses reveal limits on how precisely selection data simultaneously constrain perceptual representations and the color-material weight ., These limits should guide future efforts towards understanding the multidimensional nature of object perception .

Much is known about how the visual system extracts information about individual object properties , such as color or material ., Considerably less is known about how percepts of these properties interact to form a multidimensional object representation ., We report the first quantitative analysis of how perceived color and material combine in object selection , using a task designed to reflect key aspects of how we use vision in real life ., We introduce a computational model that describes observers’ selection behavior in terms of ( 1 ) how objects are represented in an underlying subjective perceptual color-material space and ( 2 ) how differences in perceived object color and material combine to guide selection ., We find large individual differences in the degree to which observers select objects based on color relative to material: some base their selections almost entirely on color , some weight color and material nearly equally , and others rely almost entirely on material ., A fine-grained analysis clarifies the limits on how precisely selection data may be leveraged to simultaneously understand the underlying perceptual representations on one hand and how the information about perceived color and material combine on the other ., Our work provides a foundation for improving our understanding of visual computations in natural viewing .

infographics, physical mapping, statistics, social sciences, neuroscience, charts, perception, cognitive psychology, mathematics, materials science, vision, molecular biology techniques, research and analysis methods, computer and information sciences, gene mapping, texture, molecular biology, statistical models, psychology, data visualization, color vision, biology and life sciences, sensory perception, physical sciences, material properties, cognitive science

journal.ppat.1003593

HIV-1 Superinfection Occurs Less Frequently Than Initial Infection in a Cohort of High-Risk Kenyan Women

Development of a safe and effective prophylactic HIV vaccine remains enormously challenging , due to the viruss high diversity and our limited understanding of immune correlates of protection ., While most effective vaccines are designed to mimic natural infection and protective immune responses to it , such a template for HIV vaccine design remains elusive , since sterilizing immune responses to natural infection have not been observed ., A priority of HIV vaccine development is , therefore , to identify settings where natural infection elicits some immune functions desired in a vaccine ., For example , HIV-infected individuals who spontaneously control viral replication have provided insights into immune mechanisms of HIV control 1 ., However , models where the response , rather than delaying disease , prevents infection – the ultimate goal of a prophylactic vaccine – remain less well characterized ., Studies of superinfection ( reinfection from a different partner ) provide a unique model in which to investigate the impact of pre-existing responses on susceptibility to infection by diverse circulating viral variants , which include multiple subtypes with up to 30% sequence variation ., HIV superinfection has been reported in a number of settings 2–13 , implying that HIV acquisition can occur despite the immune response to initial infection ., However , it remains an open question whether pre-existing infection affords some protection from superinfection , and individuals who do become superinfected are a select subset deficient in a particular aspect of immunity ., Published estimates of superinfection incidence vary from no identified cases 1 , 14–16 to rates roughly similar to initial infection 2–13 , 17 , 18 ., These discrepancies are largely explained by differences in participant inclusion criteria and study design ., The studies that have directly compared initial and superinfection incidence have had limited statistical power due to cohort size 5 , 12 , 17 , 18 or number of cases of superinfection identified 3 , 8 ., Additionally , methods used to identify superinfection have evolved ., Superinfection is most reliably detected in longitudinal samples by the presence of a single viral clade initially followed by introduction of a second phylogenetically distinct clade 19 ., Detection sensitivity is dependent on the number of genomic regions analyzed 12 , as well as sequencing depth 20 ., Until recently , sequences were obtained by limiting dilution amplification and Sanger sequencing 5 , 6 , 12 , 17 , which limits detection to cases where the second virus is relatively abundant ., The development of next generation sequencing ( NGS ) has enabled higher-throughput , deeper sequencing of large cohorts 20 , 21 ., To date , the largest study to examine the rate of superinfection in a prospective seroincident cohort was a NGS screen by Redd et al . of 149 individuals in which 7 cases were identified 8 ., No statistically significant difference was found between the incidences of initial infection and superinfection , though the relatively small number of cases may have resulted in limited statistical power ., A greater number of cases was found in a high-risk cohort in Mombasa , Kenya , with 12 cases of 56 women screened 5 , 12 , 17 ., However , this study used Sanger sequencing to sample ∼7 clones per sample , which could miss lower frequency variants , and was not powered to compare incidences ., In the present study , we developed a NGS method for identification of superinfection , and used it to screen 129 women in the same Mombasa cohort , including those classified as singly infected in the prior study ., We identified 9 additional cases of superinfection , for a total of 21 cases in this cohort ., These combined data enabled comparison of the incidence rates of initial infection and superinfection ., In order to conduct a sensitive , high-throughput screen for superinfection in the Mombasa cohort , we developed a pipeline for amplification , next-generation sequencing ( NGS ) , data cleaning , and phylogenetic and sequence diversity analysis of longitudinal plasma RNA ( Fig . 1 ) ., One-hundred thirty-two women met our selection criteria for the NGS superinfection screen , with a median follow-up time of 2046 days ( IQR 1265–2848 ) ., We successfully amplified gag , pol and env at three timepoints in 115 women and at least two genomic regions in at least the first and last timepoints in 129 ., The remaining 3 women were dropped from analysis ., In total , ∼1 . 7 million raw sequencing reads were obtained , with ∼1 . 25 million passing quality filtering: a median of 901 per amplicon per sample ., Women were considered putative superinfection cases if the posterior probability of monophyly supported single infection at the earliest studied timepoint followed by introduction of a distinct viral clade and increased viral diversity consistent with that seen in simulated dual infection ( Fig . 1e&f ) ., Putative cases of superinfection were confirmed and their timing specified by analyzing intervening timepoints ., Nine cases of superinfection were detected and their timing specified ., One case of suspected dual infection was detected , in which two clades were detected at the earliest sample analyzed ( 60 days post-infection ( dpi ) ) and throughout infection ( data not shown ) ., Example data from two cases of superinfection are summarized in Figures 2 and 3 ., Initial screening of subject QD151 ( Fig ., 2 ) showed monophyletic subtype A infection at 39 dpi and two subtype A clades in all three genes at 938 and 1701 dpi ., In subsequent analysis of intervening timepoints the second clade was first detectable at 801 dpi ( Fig . 2a ) ., At this time , pairwise distance increased sharply , for example in gag from 0 . 27% at 241 dpi , to 12 . 75% at 801 dpi ( Fig . 2b ) , into the range observed in simulated dual infections ., These observations supported introduction of a second subtype A variant between 241 and 801 dpi ., The initial clade was no longer detectable in pol at 1701 dpi , suggestive of a genomic recombination event ( Fig . 2c ) ., Similarly , subject QB210 ( Fig ., 3 ) showed initially monophyletic infection with a subtype A/D virus , followed by introduction of a subtype C/D virus at 163 dpi , evidenced by polyphyly and a shift in pairwise distance ( >10% ) in all 3 genes ( Fig . 3a and 3b ) ., In intervening timepoints , the second variant could be detected in all genes at 163–170 dpi , but was undetectable in gag and pol after 170 dpi , indicating recombination ( Fig . 3c ) ., Characteristics of the 9 new cases of superinfection are summarized in Table 1 and Figure S2 ., In all but two cases the superinfecting variant was detected in all 3 amplicons in at least one timepoint ., In all cases , the superinfecting variant was detected at multiple timepoints in at least one amplicon ., In one case ( QC369 ) , the initial variant became undetectable in any amplicon following superinfection , suggesting it was replaced , to our detection limit , by the superinfecting variant ., Both variants were detected at two timepoints each , the initial variant at 17 dpi and 28 dpi , and the superinfecting variant at 143 dpi and 451 dpi ( Fig . S2 ) , indicating this result was not due to contamination ., Further , the possibility of sample mix-up was excluded by HLA-typing ( data not shown ) ., As illustrated in Figures 2 , 3 and S2 , in the other 8 cases , variants were intermittently detected in different amplicons at different times , suggestive of genomic recombination and dynamic turnover of the circulating viral population ., Combining the data here with those from previous studies in the Mombasa cohort 5 , 12 , 17 , a total of 146 women were examined for superinfection: 90 were tested using NGS , 39 using both NGS and Sanger sequencing , and 17 using only Sanger sequencing ., Among the 39 women previously identified as singly infected by Sanger sequencing and tested by NGS here , no new cases of superinfection were identified , suggesting older methods were sensitive enough to detect superinfection ., Twenty-one cases of superinfection were confirmed based on detection of the superinfecting virus in two or more samples ., The timing windows of all 21 superinfection events are summarized in Figure 4 and Table S2 ., The midpoint of the timing window of the 9 new cases ranged from 81 to 1041 dpi , with 6 occurring within the first year of infection ., The window of superinfection events was defined to a median of within 127 days , with window sizes of 90 to 1253 days ., Timing of all 21 cases ranged from 63 to 1895 dpi , defined to a median of within 146 days ., We detected both inter-subtype and intra-subtype superinfections ., In 6 of 9 cases identified by NGS , the superinfecting variant was the same subtype as the initial variant in every gene where both were detected ., In all 9 cases , the variants were the same subtype in the env amplicon ( Table 1 ) ., Among all 21 cases of superinfection ( Table S2 ) , the majority of superinfection events we detected were intrasubtype , regardless of genomic region: 53 . 8% were intrasubtype based on gag sequence , 62 . 5% based on pol , and 70 . 6% based on env ., We further investigated the possibility of a bias in sequence similarity of superinfecting variants to initial variants by analyzing amino acid diversity ., We compared the pairwise amino acid distance between initial and superinfecting variants within each superinfection case to the distance that would be expected by chance ., The latter was modeled by simulated mixtures of sequences from all possible pairs of singly infected individuals in the Mombasa cohort ( Fig . 5 ) ., Using NGS data from the 9 superinfection cases and 120 singly infected women screened here , we found no significant differences between the sequence similarity within superinfected individuals and that expected by chance ( Fig . 5a ) ., Including Sanger sequencing data from the additional 12 superinfected women previously screened yielded a similar result ( Fig . 5b ) The incidence of superinfection among women who were screened was compared to the incidence of initial infection in the entire cohort at risk ., Only incident HIV infections ( occurring after enrollment in the cohort ) were included ., Fourteen women who were seronegative but HIV RNA positive at enrollment were excluded for this reason ., Seven of these had been screened for superinfection , and one was found to be superinfected , which mirrors the frequency of superinfection observed in the entire group ., The individual with evidence of dual infection at the earliest timepoint was also excluded , since we were unable to distinguish coinfection from superinfection ., After exclusions , 1910 women were at risk of initial infection , contributing 5124 person-years , and 138 women were screened for superinfection , contributing 764py following first infection ., There were 295 initial infections , giving a crude incidence rate of 5 . 7 per 100pys , and 20 superinfections , giving a crude incidence rate of 2 . 61 per 100 pys ., The incidence of superinfection and initial infection over time is summarized in Figure 6 ., We used Andersen-Gill proportional hazards analysis to generate a hazard ratio ( HR ) relating the incidence of superinfection to that of initial infection ., The unadjusted HR for this comparison was 0 . 49 ( CI 0 . 31–0 . 76 , p\u200a=\u200a0 . 0018 ) ., Variables previously shown to influence HIV exposure risk in this cohort 22 , 23 were included as adjustments in the model ( summarized in Table 2 ) ., These included self-reported sexual risk behavior , place of work , hormonal contraceptive use , genital tract infections , years in sexwork , age at first sex , total follow-up time in the cohort and calendar year ., The HR for superinfection compared to initial infection , adjusted for these variables , was 0 . 47 ( CI 0 . 29–0 . 75 , p\u200a=\u200a0 . 0019 ) ., Since proportional hazards analysis is based on time to infection and the precision with which superinfection timing was determined varied between cases , we performed sensitivity analyses setting infection timing for all cases to the start or midpoint of the timing windows rather than the end , as done for the above analysis ., In both of these analyses , significant differences in incidence were also observed: setting infection timing to the start of the windows , the adjusted HR was 0 . 33 ( CI 0 . 18–0 . 58 , p\u200a=\u200a0 . 00012 ) ; using the window midpoints , the adjusted HR was 0 . 39 ( CI 0 . 23–0 . 63 , p\u200a=\u200a0 . 00016 ) ., We assessed whether the risk of superinfection varied with time since initial infection by dividing our data into infection events occurring early or late in follow-up and estimating the HR , as above , in each subset ., We found that within the first 6 months at risk , the incidence rates of initial and superinfection did not differ significantly ( adjusted HR 0 . 73 , p\u200a=\u200a0 . 51 ) , whereas after 6 months the rate of superinfection was lower than that of initial infection ( adjusted HR 0 . 40 , p\u200a=\u200a0 . 0017 ) ., A similar result was observed when considering events within or beyond one year at risk: within the first year , the incidence rates of initial and superinfection did not differ significantly ( adjusted HR 0 . 54 , p\u200a=\u200a0 . 14 ) , but after one year the rate of superinfection was significantly lower ( adjusted HR 0 . 43 , p\u200a=\u200a0 . 0059 ) ., Sensitivity analyses setting infection time to the start and midpoint of the timing windows as above reproduced the same results ( data not shown ) ., We noted that the previous screens in the cohort appeared to detect a higher frequency of superinfection than the NGS screen ( 12 cases of 56 women screened , compared with 9 cases of 90 ) , with a greater fraction of the events occurring later after initial infection ( Fig . 4 ) ., Since the NGS screen spanned later years in the cohort than the previous studies , such a difference could be due to the known decline in infection risk in the cohort over calendar time 22 , 23 ., However , the numbers of events are small when the datasets are considered separately and the difference both in superinfection incidence rate and post-infection timing between the two studies was not statistically significant ( data not shown ) ., In this study we used NGS to screen for superinfection in 129 high-risk women and identified 9 cases of superinfection ., Combined with previous studies5 , 12 , 17 , a total of 21 cases of superinfection were detected among 146 women screened in this cohort ., There was a statistically significant difference between the incidence of superinfection ( 2 . 61 per 100pys ) and initial infection ( 5 . 75 per 100 pys ) , with a hazard ratio of 0 . 47 after adjusting for potential confounding factors ., This suggests that HIV infection provides partial protection from subsequent infection ., The relatively large size of this cohort and high number of superinfection cases enabled us to detect for the first time a statistically significant difference between the incidence of initial infection and superinfection ., This possibility has been proposed previously , though the studies were not designed and/or powered to detect a difference 17 , 18 ., In the largest incidence study prior to the present study , Redd et al . screened a comparable number of individuals ( 149 ) in a lower-risk cohort and identified 7 cases of superinfection ., The incidence of superinfection was not found to differ significantly from initial infection , but there was a trend for lower incidence of superinfection when controlling for baseline sociodemographic differences between the groups at risk of initial and superinfection ., Analysis of our data using the same methods as Redd et al . – Poisson regression with propensity score matching 8 – was consistent with the results of our Andersen-Gill analysis , showing a significant difference in incidence , with an estimated incidence ratio of 0 . 48 ( p\u200a=\u200a0 . 011 ) comparing superinfection to initial infection ., In addition to sample size , two strengths of our incidence analysis were our specification of infection timing to within a few months on average and our comparison of initial and superinfection risk within the same cohort ., These enabled us to adjust for the same potential confounding factors in both the initial infection and the superinfection risk sets , using frequently collected time-varying covariate data ., Particularly important , given the sequential nature of superinfection , was adjustment for calendar year to control for decline in infection risk in the cohort over time ., The distributions of initial and superinfection events over calendar time were similar ( Fig . S3 ) , suggesting community-level changes over time did not severely bias our analysis ., The ∼two-fold reduction we found in the incidence of superinfection has a number of possible interpretations ., First , it may indicate that the adaptive immune response elicited by initial infection provides partial protection from second infection ., If this were the case , superinfection might preferentially occur early in infection , before the response has matured 2 , 13 , 24 ., In support of this idea , we found that , although superinfection occurred throughout the course of first infection , the incidence of superinfection was significantly lower than initial infection after the first 6 months of infection , but not earlier ., This suggests that susceptibility to superinfection decreased over time , coincident with broadening and strengthening of HIV-specific immunity ., Indeed , this has been suggested by two earlier studies , each documenting three cases of superinfection that occurred within the first year after initial infection 3 , 18 ., If the difference in incidence we observed is due to a partially protective adaptive immune response , we would anticipate superinfection would preferentially occur with more distantly related viruses , more likely to escape the response ., Using viral subtype and pairwise amino acid distance as surrogate measures of antigenic distance , our data provided no evidence of this effect ., The majority of the 21 superinfection events we detected were intrasubtype , and the proportion of subtype A , C and D viral sequences was similar for the initial and superinfecting viruses , consistent with the subtype distribution in this cohort 25 ., The pairwise distance between initial and superinfecting variants was no higher than the distribution of distances between random pairs of singly-infected individuals from the Mombasa cohort ., This may potentially be explained by limited sample size or insufficient simultaneously circulating subtypes ., It also may be that sequence relatedness is a poor indicator of susceptibility to the immune response or the genome regions we analyzed are not critical antigenic determinants of protection ., Alternatively , it is possible that protective immune responses are not driving the protective effect we observed ., Another potential explanation for the lower risk of superinfection is that HIV infection itself may reduce infection risk by depleting permissive target cells ., On the other hand , chronic immune activation and immunodeficiency following HIV infection could increase susceptibility , potentially blunting protective effects 26 ., Thus , there may be a complex interplay of biological factors impacting HIV risk in an HIV-positive individual ., So far , studies of immune correlates of superinfection have yielded variable results – some suggesting neutralizing antibody deficits in superinfection 27 , 28 , while others , including studies in the Mombasa cohort , detected no differences in antibody 29 , 30 or cellular 31 responses ., A major challenge in these studies has been the identification and analysis of large enough numbers of superinfection cases: the small sample sizes in studies to date ( three to twelve superinfected individuals ) would restrict detection to only very large effects ., Small sample size is just one factor that has made detecting immune deficits associated with superinfection challenging and contributed to variable results among studies ., There has also been variation among published studies in the control groups used for comparison , including the time at which the response was analyzed relative to the time of superinfection and initial infection ., Given the dynamic nature of the immune response , sample timing could impact measures in both controls and cases ., Furthermore , precision in the estimated timing of superinfection varies between studies , and between cases , providing an additional variable ., Divergent findings between studies may also reflect differences in the assays used and subtleties in the immune parameters they capture ., Our finding of lower risk of superinfection than initial infection provides greater impetus for larger-scale comprehensive analysis of multiple immune mechanisms , including both those analyzed in the smaller studies to date and , perhaps of more interest , those not characterized in prior studies ., If the discrepancies in earlier studies reflect the fact that multiple immune parameters are at play , then examining a variety of immune responses in the same individuals in a larger cohort may be needed to define responses that contribute to HIV susceptibility following initial infection ., Like all studies , the study presented here has a number of limitations ., Firstly , while our screening methods are among the most sensitive developed , it remains possible that some cases of superinfection were missed ., In particular , reinfection by the same source partner is not captured by any existing methods ., Additionally , our specification of the timing of superinfection was limited by the samples available to us ., While follow-up was generally frequent in this study population , there were six superinfection cases where sample availability limited our ability to define the time of superinfection to within a one-year period ., This uncertainty in superinfection timing did not affect our findings , as we found that whether we assumed in the incidence analysis that the true timing of superinfection was at the start , midpoint or end of the timing window , the results indicated that the incidence of superinfection was significantly lower than that of initial infection ., Finally , as in all observational studies , residual confounding of our incidence estimate by behavioral changes and sexual network-level factors not measured or accounted for in our analyses remains a possibility ., However , the fact that we compared initial and superinfection risk within the same cohort and collected covariate data at frequent intervals enabled us to minimize this issue to an extent not possible in previous studies ., This study provides the first robust evidence that HIV infection reduces the risk of subsequent infection ., The underlying mechanism remains unclear , but this finding prompts exploration of correlates of protection from HIV in high-risk individuals who continue to be exposed after first infection ., Furthermore , this study reinforces that superinfection occurs at a considerable rate , calling for studies of its impact on the clinical progression , transmission , and epidemiology of HIV ., The study was approved by the ethical review committees of the University of Nairobi , the University of Washington and the Fred Hutchinson Cancer Research Center ., Written informed consent was obtained from all participants ., Seronegative women in Mombasa , Kenya , attended monthly visits , at which clinical examinations , interviews and sample collection took place , as previously described 22 ., Following seroconversion , sample collection took place quarterly ., Individuals were selected for superinfection screening based on sample availability <6 months and >2 years post-initial HIV infection , and an approximately equally spaced intervening sample ., Within these limitations , samples with maximal plasma viral load , >1000 copies/ml , and prior to initiation of antiretroviral therapy were selected ., Thirty-nine of 44 women previously screened for superinfection by Sanger sequencing and identified as singly infected 5 , 12 , 17 were rescreened; the remaining 5 women did not have adequate samples available ., HIV virions were isolated from heparinized plasma using the μMACS VitalVirus HIV Isolation kit ( Miltenyi Biotec ) and viral RNA extracted from 140–420 µl , depending on viral load , using the Qiamp viral RNA Mini kit ( Qiagen ) ., Nested RT-PCR of ∼500 bp in gag , pol and env was conducted in duplicate ( see Table S1 ) ., RNA input into each reaction was normalized to 3000 viral genomes according to plasma viral load , or the maximum possible where viral load was too low ., RT-PCRs for the three genes were multiplexed ., Nested PCR reactions were carried out separately for each region with primers containing adaptors for Roche 454 sequencing and a unique 8 bp barcode sequence to identify each sample ., PCR products were purified using AMPure XP PCR purification beads ( Agencourt ) and quantified using the Qubit dsDNA HS assay ( Invitrogen ) ., PCR products were sequenced on the Roche 454 GS-Junior or GS-FLX titanium platform ., Where initial sequencing suggested superinfection ( see below ) , timing was inferred by sequencing intervening timepoints ., Sequences are available upon request from the authors ., 454 sequences were error-corrected using AmpliconNoise 32 ., Chimeric sequences were identified and removed using UCHIME 33 ., Cross-contamination between samples sequenced together and contamination by other lab samples was identified by all-against-all BLAST against a local database of published HIV sequences and sequences from the same sequencing run ., Sequences with high identity hits to known laboratory stains or other samples from the same sequencing run were removed ., Sequences with abundance <5 reads or 0 . 5% of the sample , whichever was higher , were excluded from further analyses as lower abundance variants were not reproducibly detected in repeated deeper sequencing of two selected samples where rare variants formed a distinct phylogenetic clade ., An amplicon-specific profile HMM was created from an alignment of representative sequences from multiple subtypes ., For each subject and amplicon , 20 reference sequences were selected by placing 454 reads on a tree of candidate reference sequences 34 and minimizing the average distance to the closest leaf 35 ., These reference sequences , representatives from subtypes common to the region , and 454 reads were aligned to the HMM using hmmalign 36 and non-consensus columns removed ., Any sequences <200 bp long after alignment and trimming were removed ., We used BEAST 37 to calculate a posterior probability of monophyly for the sequences ., A posterior sample of trees was obtained using a strict molecular clock , Bayesian Skyline Plot population model and the HKY substitution model ., Each MCMC chain ran 20 million iterations , sampling every 2000 , discarding the initial 25% of samples as burn-in ., Chains were assessed for convergence by examining effective sample size ( ESS ) and by visual inspection of traces of key parameters ., A strict clock was used as poor mixing was frequently observed under relaxed clock models ., BEAST runs with intermediate posterior probabilities ( 0 . 2–0 . 8 ) were manually examined for recombinant sequences and run again with putative recombinants removed ., Pairwise distances were calculated for all sequence pairs under the TN93 model using APE 38 , reporting the maximum within-subject distance ., For comparison , 95% confidence limits of pairwise distances were calculated for sequences from known single infections ( previously screened in 5 , 12 , 17 ) and simulated dual infections ., Dual infections were simulated by combining all pairs of sequences from previously screened singly infected samples ., Pairwise distances calculated from 454 sequences obtained in this study were compared to the upper bound of the 95% quantile of single infection distances , and the lower bound of the 95% quantile of simulated dual infection distances ., This pipeline was validated and refined by processing monophyletic viral isolates , known mixtures of isolates , and known cases of superinfection detected by Sanger sequencing 17 ., These methods were found to be sensitive enough to distinguish two subtype A isolates mixed at abundances of 5%∶95% genome copies in all three genomic regions , and at 1%∶99% in two of three genomic regions ( Fig . S1 ) ., Sequences were aligned as for the phylogenetic analysis ., Insertions relative to the reference alignment were removed , and sequences with <60% coverage or identified as recombinants between initial and superinfecting variants upon visual inspection were excluded ., For each case of superinfection , viral sequences were annotated as the initial strain or the superinfecting strain ., We calculated the mean Hamming distance between amino acid sequences of the superinfecting strain from the time of superinfection detection and sequences of the initial strain up to and including this time ., In calculating the mean distance , each pairwise comparison was weighted using the product of the multiplicities of the two reads ., To investigate whether these distances deviated from what would be expected by chance , an artificial set of mock superinfections was generated by combining sequences from singly infected individuals ., All pairs of singly infected individuals screened by 454 sequencing were enumerated ., In each pair , one individual was randomly chosen to be the source of the ‘initial’ virus in the simulated superinfection ., A time of ‘superinfection’ was chosen randomly from the available sampled timepoints and sequences from all timepoints up to and including this time were used for analysis ., The other individual in the pair acted as the source of the ‘superinfecting’ virus ., A time of ‘transmission’ was chosen randomly from the available sampled timepoints and sequences from this timepoint were used ., Mean distances within pairs were calculated as above ., The analysis was repeated including gag and env Sanger sequences from previously published cases 5 , 12 , 17 , trimmed to the genome region amplified for NGS , and given unit weight ., A two-sample Wilcoxon test was used to test for a difference between the distances observed in true superinfections and those simulated in mock superinfections ., Statistical analysis was performed using R ( www . r-project . org ) ., The incidences of initial and superinfection were compared by Andersen-Gill proportional hazards analysis ., The predictor was inclusion in the screen for superinfection , modeled as a time-dependent variable , and the outcome was time to HIV infection ( initial and super ) ., Timing of infection events for the incidence analysis was set to the study visit of their detection ( for initial infection events the visit after inferred infection timing; for superinfection events , the time at which the superinfecting virus was first detected ) ., Individuals who were HIV infected but not screened for superinfection were censored after acquisition of initial infection ., Individuals who became superinfected were censored after acquisition of superinfection ., Individuals who were screened and not found to be superinfected were censored at the last timepoint screened ., Since samples after initiation of antiretroviral treatment were excluded from superinfection screening , no follow-up after treatment initiation was included ., The model was adjusted for time-varying variables at each visit: calendar year , age , years in sexwork , number of weekly sexual partners , number of weekly unprotected sex acts , hormonal contraceptive use in the prior 70 days and any genital tract

Introduction, Results, Discussion, Materials and Methods

HIV superinfection ( reinfection ) has been reported in several settings , but no study has been designed and powered to rigorously compare its incidence to that of initial infection ., Determining whether HIV infection reduces the risk of superinfection is critical to understanding whether an immune response to natural HIV infection is protective ., This study compares the incidence of initial infection and superinfection in a prospective seroincident cohort of high-risk women in Mombasa , Kenya ., A next-generation sequencing-based pipeline was developed to screen 129 women for superinfection ., Longitudinal plasma samples at <6 months , >2 years and one intervening time after initial HIV infection were analyzed ., Amplicons in three genome regions were sequenced and a median of 901 sequences obtained per gene per timepoint ., Phylogenetic evidence of polyphyly , confirmed by pairwise distance analysis , defined superinfection ., Superinfection timing was determined by sequencing virus from intervening timepoints ., These data were combined with published data from 17 additional women in the same cohort , totaling 146 women screened ., Twenty-one cases of superinfection were identified for an estimated incidence rate of 2 . 61 per 100 person-years ( pys ) ., The incidence rate of initial infection among 1910 women in the same cohort was 5 . 75 per 100pys ., Andersen-Gill proportional hazards models were used to compare incidences , adjusting for covariates known to influence HIV susceptibility in this cohort ., Superinfection incidence was significantly lower than initial infection incidence , with a hazard ratio of 0 . 47 ( CI 0 . 29–0 . 75 , p\u200a=\u200a0 . 0019 ) ., This lower incidence of superinfection was only observed >6 months after initial infection ., This is the first adequately powered study to report that HIV infection reduces the risk of reinfection , raising the possibility that immune responses to natural infection are partially protective ., The observation that superinfection risk changes with time implies a window of protection that coincides with the maturation of HIV-specific immunity .

HIV-infected individuals with continued exposure are at risk of acquiring a second infection , a process known as superinfection ., Superinfection has been reported in various at-risk populations , but how frequently it occurs remains unclear ., Determining the frequency of superinfection compared with initial infection can help clarify whether the immune response developed against HIV can protect from reinfection – critical information for understanding whether such responses should guide HIV vaccine design ., In this study , we developed a sensitive high-throughput method to identify superinfection and used this to conduct a screen for superinfection in 146 women in a high-risk cohort ., This enabled us to determine if first HIV infection affects the risk of second infection by comparing the incidence of superinfection in this group to the incidence of initial infection in 1910 women in the larger cohort ., We found that the incidence of superinfection was approximately half that of initial infection after controlling for behavioral and clinical differences that might affect infection risk ., These results suggest that the immune response elicited in natural HIV infection may provide partial protection against subsequent infection and indicate the setting of superinfection may shed light on the features of a protective immune response and inform vaccine design .

immunodeficiency viruses, medicine, virology, epidemiology, biology, microbiology

journal.pcbi.1000307

Probabilistic Interaction Network of Evidence Algorithm and its Application to Complete Labeling of Peak Lists from Protein NMR Spectroscopy

The usual approach to the solution of the problem of assigning labels to subsets of peaks ( spin subsystems ) assembled from multiple sets of noisy spectra is to collect a number of multidimensional , multinuclear datasets ., After converting the time domain data to frequency domain spectra by Fourier transformation , peaks are picked from each spectrum for analysis ., Methods have been developed for automated peak picking or global analysis of spectra to yield models consisting of peaks with known intensity , frequency , phase , and decay rate or linewidth 7 , 8 ., In the ideal case , the resulting peak-lists identify combinatorial subsets of two or more covalently bonded nuclei by their respective frequencies ( Figure 2 ) ., These subsets must be “assembled” in a coherent way to “best” correspond to specific atoms in the amino acid sequence of the protein ., In practice , peak lists do not report on all nuclei ( because some peaks are missing ) , and “noise peaks” ( peaks incorrectly reported as true peaks ) are commonplace ., In the examples analyzed here ( Table 1 ) , the level of missing peaks varied between 9% and 38% , while the level of noise peaks varied between 10% and 135% ., The large number of false positives as well as false negatives typically present in the data result in an explosion of “ambiguities” during the assembly of subsets ., A common feature among prior approaches has been to divide the assignment of labels into a sequence of discrete steps and to apply varying methods at each step ., These steps typically include an “assignment step” 9–12 , a secondary structure determination step 13–15 , and a “validation step” 16 ., The validation step , in which a discrete reliability measure indicates the possible presence of outliers , misassignments , or abnormal backbone chemical shift values , is sometimes omitted ., Other steps can be added , or steps can be split further into simpler tasks ., For example , backbone and side chain assignments frequently are carried out sequentially as separate processes ., Some approaches to sequence-specific assignment rely on a substantially reduced combinatorial set of input data by assuming a prior subset selection , e . g . , prior spin system assembly 17 , 18 ., The specification of conformational states can be added as yet another labeling step ., For example , backbone dihedral angles can be specified on a grid ( e . g . , 30° intervals ) as determined from chemical shifts 19 , coupling constants and/or NOEs 20 , or reduced dipolar couplings 21 ., The NMR assignment problem has been highly researched , and is most naturally formulated as a combinatorial optimization problem , which can be subsequently solved using a variety of algorithms ., A 2004 review listed on the order of 100 algorithms and software packages 22 , and additional approaches are given in a 2008 review 23 ., Prior methods have included stochastic approaches , such as simulated annealing/Monte Carlo algorithms 24–26 , genetic algorithms 27 , exhaustive search algorithms 17 , 28–30 , heuristic comparison to predicted chemical shifts derived from homologous proteins 31 , heuristic best-first algorithms 32–34 , and constraint-based expert system that use heuristic best-first mapping algorithm 35 ., Of these , the most established , as judged from BMRB entries that cite the assignment software packages used , are Autoassign 10 and GARANT 27 ., Similarly , a wide range of methods have been used to predict the protein secondary structural elements that play an important role in classifying proteins 36 , 37 ., Prior approaches to assigning a secondary structure label to each residue of the protein have included the Δδ method 38 , the chemical shift index method 14 , a database approach ( TALOS ) 19 , an empirical probability-based method 39 , a supervised machine learning approach 40 , and a probabilistic approach that utilizes a local statistical potential to combine predictive potentials derived from the sequence and chemical shifts 13 ., Recently , a fully automated approach to protein structure determination , FLYA , has been described that pipelines the standard steps from NMR spectra to structure and utilizes GARANT as the assignment engine 41 ., The FLYA approach demonstrates the benefits of making use of information from each step in an iterative fashion to achieve a high number of backbone and side chain assignments ., Our goal is to implement a comprehensive approach that utilizes a network model rather than a pipeline model and relies on a probabilistic analysis for the results ., We reformulate the combinatorial optimization problem whereby each labeling configuration in the ensemble has an associated but unknown non-vanishing probability ., The PINE algorithm enables full integration of information from disparate steps to achieve a probabilistic analysis ., The use of probabilities provides the means for sharing and refining incomplete information among the current standard steps , or steps introduced by future developments ., In addition , probabilistic analysis deals directly with the multiple minima problem that arises in cases where the data does not support a single optimal and self-consistent state ., A common example is a protein that populates two stable conformational states ., The PINE-NMR package described here represents a first step in approaching the goal of a full probabilistic approach to protein NMR spectroscopy ., PINE-NMR accepts as input the sequence of the protein plus peak lists derived from one or more NMR experiments chosen by the user from an extensive list of possibilities ., PINE-NMR provides as output a probabilistic assignment of backbone and aliphatic side chain chemical shifts and the secondary structure of the protein ., At the same time , it identifies , verifies , and , if needed , rectifies , problems related to chemical shift referencing or the consistency of assignments with determined secondary structure ., PINE-NMR can make use of prior information derived independently by other means , such as selective labeling patterns or spin system assignments ., In principle , the networked model of PINE-NMR is extensible in both directions within the pipeline for protein structure determination ( Figure 1 ) : it can be combined with adaptive data collection at the front or with three-dimensional structure determination at the back end ., Such extensions should lead to a rapid and fully automated approach to NMR structure determination that would yield the structure most consistent with all available data and with confidence limits on atom positions explicitly represented ., In addition to its application to NMR spectroscopy , the PINE approach should be applicable to the unbiased classification of biological data in other domains of interest , such as systems biology , in which data of various types need to be integrated: genomics ( DNA chips ) , proteomics ( MS analysis of proteins ) , and metabolomics ( GC-MS , LC-MS , and NMR ) data collected as a function of time and environmental variables ., The fundamental idea of PINE is to embed the original assignment problem into a higher dimensional setting and to use empirically estimated compatibility ( or similarity ) conditions to iteratively arrive at an internally coherent labeling state ., These conditions are embodied in the form of a parameterized Hamiltonian ( energy function ) that evolves at each iteration step ., In the quasi-stationary regime , this construction yields clusters , defined as subsets of chemical shift data with assigned labels ., The clusters have strong intra-cluster links and highly localized inter-cluster couplings ., We view each possible cluster of related experimental data in the domain as a “site” that is to be potentially labeled ., More specifically , our goal is to discover ( learn ) the map f that relates the “domain” ( set of subsets of data ) to the “codomain” ( set of subsets of labels ) :where X\u200a=\u200ax1 , x2 , … , xm is the set of data values available from all experiments , and L\u200a=\u200aL1 , L2 , … , Ln is the set of labels associated to the chemical shifts ., At first it may appear that this map is trivial , because one protein has precisely one set of correct chemical shifts ., However , breaks in the backbone sequential data , incompleteness of experimental peak lists , and the presence of many noise peaks renders the discovery of a deterministic one-to-one map to the sequential labels unpromising ., Rather than discovering a single map , we opt to find a set of maps , each with its associated probability ., More directly , we choose to associate subsets of labels from the list L to subsets of data from the list X , each with a commensurate probability: In order to formulate the computational problem , we require that the labels for data values satisfy constraints that arise from the system of neighborhoods built around each data value ., The system of neighborhoods is a dynamic state variable that co-evolves with the probability values ., We assign an initial set of labels , L , with associated weights to each input data point , S ( e . g . , chemical shift ) and introduce a measure of similarity based on distances between “neighboring points” ( Figure 3 ) ., Typically , in our starting configuration , the possible labels for each data value far exceed the number of sites ., The set of labels contains the “null” label to allow for the case where a data element cannot be labeled ., The approach used to measure the global compatibility or support for the specific labeling of site S at iteration step m is to aggregate the compatibilities over versions of individual evidences by applying a variation of the belief propagation algorithm 42 ., The evidence for assignment is weighted by the probability of each “neighbor” being correct , and the probabilities at stage m can be updated by replacing them by the new weight configuration state ., As probabilities evolve , the information content of changing configurations is monitored for the optimally “informative” state ., The resulting model is analogous to the random cluster Fortuin and Kasteleyn ( FK ) model 43 ., In practice , a straightforward implementation leads to densely connected networks with noisy weights and no principled way to control the iteration steps ., To implement the intuitively appealing ideas presented above that are designed to find the optimal state in the form of marginal probabilities , we have devised an iterative approach that utilizes topology selection followed by a variation of belief propagation algorithm 42 and subsequent adjustment of initial weights and topology ., This topology selection step plays a key role in achieving robust and computationally efficient results ., We proceed by analogy to FK 43 ., Let G\u200a= ( V , E ) be any general graph , with e∈E an edge in G , and ν∈V a vertex ., The set of assignments ( or labels ) for each vertex is designated by 1 , 2 , … , q ., The “configuration energy” of this system is encoded in the partition function: ( 1 ) In this formula , the outside sum is performed over the configuration states of the system represented by the map λ , and the inside product measures the compatibility of the vertex labels joined by the edge e ., Each edge is weighted by the factor and has end-point vertices , and δ is the compatibility measure of end-point vertices configuration ., By defining and , Eq 1 can be rewritten as: ( 2 ) In the setting of statistical physics , the Boltzmann weight of a configuration is , where H ( the sum in the exponential ) represents the energy of the configuration and β is a parameter called the inverse temperature ., Because the weights are assumed to be positive , they can be interpreted probabilistically ( after normalization by Z ) as a probability measure on the states for the graph G where N is the number of vertices ., In the standard random-cluster model , the neighborhood structure , or topology , of the graph is prescribed , and the objective is to find the ground state for a given set of weights by varying the “spin” , or labeling , configurations ., In our case , we are determining the ground state ensemble and the topology of the model at the same time ., At each iteration step i , we define Ai , a subset of the graph G , where , and evaluate the partition function for this subset ., We evolve the topology of the graph at each iteration by the addition and removal of edges and by refining the edge weights toward the optimum topology as described in the algorithm section ., A local Bayesian updating procedure updates the weights , and the local rate of change of weights is used to modify the corresponding local topology of the graph ., On the subsequent iteration , our algorithm reintegrates these local modifications in the context of the entire network and attempts to establish a new quasi-stationary state ., The algorithm must address two critical challenges ., The data that describe edge weights and states in Eq 2 are derived from empirical relationships that involve noisy data , and , therefore , a straightforward deterministic search of the resulting combinatorial space would be infeasible ., In addition , the computational complexity of the resulting problem grows rapidly with the number of labels and the topology of the graph; thus , a suitable starting and evolving representation of the topology , and a corresponding approximation algorithm is the key to obtaining a robust solution to this problem ., The probabilistic construction used in PINE-NMR belongs to the general class of graphical models in which dependencies among random variables are constructed ahead of the inference task ., In cases where the graph of dependencies is acyclic , there are powerful and efficient algorithms that correctly maximize the marginal probabilities through collecting messages from all leaf nodes at a root node 44 ., When the graph is not acyclic , current algorithms for graphs with cycles often reach oscillatory states , converge to local maxima , or achieve incorrect marginals due to computational difficulties ., Approaches have been described in the literature for dealing with a single loop condition 45 or for converging under alternative free energy approximations 46 , 47 ., “Tree-based reparameterization” algorithms 48 have been described as a general approach that iteratively reparameterizes the distributions without changing them on the subtrees in the original graph ., These algorithms , which are geared toward addressing the approximation of marginals in the presence of loops , represent trade-offs among robustness , accuracy , computational speed , and efficiency of implementation ., Our modification provides a simple extension that can be described as an adaptive form of coarse-to-fine approximation ., We start with a “coarser topology” and explore more refined factorizations of the probability distribution and look for stable fixed points ., In our adaptive approach , the extension of the state space ( embodied in the algorithm ) plays a critical role ., In intuitive terms , the additional degrees of freedom ( null states ) provide “room for change” for existing distributions as the topology is being refined ., The internal working of the stepwise factorization of the probability distribution requires a coarse estimate on the initial threshold that reduces the connectivity degree of the graph ., In our case , this approximation is arrived at using a combination of theory and empirical investigation ., Figure 4 presents an overview of the probabilistic network implemented in PINE-NMR ., Sets of probabilistic influence sub-networks are combined into a larger influence network , and each sub-network may have its own computational model used to perform the inference task ., The entire probabilistic network is constructed by considering the conditional dependencies of the sub-networks ., The actual implementation of PINE-NMR entails a fairly complicated network with more than 30 , 000 lines of code in Matlab and other supporting scripting language ., A descriptive and stepwise version is given below ., 1 . Read input data and check for errors ., If errors are found , report errors and abort ., 2 . Align the 1H , 15N , and 13C dimensions of all spectra independently ., 3 . Generate spin systems ( Figure 5 ) ., 4 . Estimate the b factor and c factor , which are the measures of data quality defined as follows:In calculating any of the above formulas , only the fields with choices are considered ., For example if none of the experiments provided by the user has HA information , HA fields are not used in the calculation ., 5 . If ( b<0 . 4 or c factor<0 . 2 ) # comment: Report low data quality to the user and stop ., The low data quality check can be manually overridden through user requests ., However , low “quality factors” are strong indicators of “highly incomplete” data and the web service discourages the use of results from low quality data ., 6 . Otherwise , set K\u200a=\u200a0 ( iteration counter ) ., Repeat: 7 . K\u200a=\u200aK+1; ( iteration counter ) ., 8 . Triplet amino acid typing: 9 . Derive the backbone assignment network weights based on amino acid typing scoring , connectivity experiments , latest backbone assignment , and possible outlier detections from the last iteration ( Figure 6 ) :T is a threshold value for the connectivity score , which is defined as , c*max_connectivity_score , c is the quality factor defined in 5 , and Pk-1 ( xn, ( i ) ) is the probability of assigning xn, ( i ) to triplet residue n in the iteration k−1 ., 10 . Select the network topology; calculate the threshold for removing low-weight edges from the network based on the quality of the data , use: 11 . Apply the belief propagation algorithm 42 to find the marginal probabilities Pkn ( xn, ( j ) ) of assigning triplet spin systems xn, ( j ) to triplet ( tripeptide ) residues n ., 12 . Given the marginal probabilities of the triplet residue assignments , derive the probabilistic assignment of the individual backbone atoms ., 13 . Detect and remove the outliers in the backbone assignments 16 ., 14 . Derive the secondary structure of each amino acid based on the formula: ( 5 ) pn ( s|xn, ( j ) ) is the probability of residue n to be in the secondary structure state s given the assignment xn, ( j ) derived from the method described in 13 , and Pkn ( xn, ( j ) ) is the assignment probability of triplet residue with the center residue n , to triplet spin system xn, ( j ) ., The summation is over all the possible choices of tripeptides in the protein sequence ., 15 . If no convergence , probabilities are the average probability of last three iterations ., “No convergence” indicates the presence of “nearby” local minima ., 16 . For every amino acid , generate an energetic model network and apply the Belief Propagation 42 to derive final probabilistic side chain assignments as described in supplementary material Protocol S1 ., 17 . Report the final probabilistic assignments: backbone , side chain , secondary structure prediction , and possible outliers ., The output can be specified to conform to variety of formats , including Xeasy , SPARKY , and NMR-STAR ( BMRB ) ., The input to PINE-NMR consists of the amino acid sequence and multiple datasets known as peak lists ( chemical shifts ) obtained separately from selected , defined NMR experiments ., The peak lists consist of sets of real-valued two-dimensional , three-dimensional , or four-dimensional vectors , denoted by lXij∈Rl l\u200a=\u200a2 , 3 , 4 ., The dimension of the data is denoted by l , the index j indicates that the observation is from the jth dataset , and the index i denotes the ith observation within the dataset ., To compare data from different experimental sets ( different, j ) that have shared subspaces ( signals from nuclei in common ) , we consider only the common subspace ., This allows us to omit the index l in subsequent formulas ., The similarity ( or nearness ) is used to build an initial system of neighborhoods ., The approximate starting value for similarity is given a probabilistic interpretation by using Eq 3 ( Basic Algorithm: 3 . a ) to compare each datum ( peak ) Xij with the reference datum ( peak ) Xmn ., The peaks in the most sensitive experiments in the dataset ( normally 15N-HSQC or HNCO ) are used as the initial reference set ., We define a common putative object , called the spin system ( Figure 6 ) , by aligning the peaks along the common dimensions and by registering them with respect to reference peaks according to Eq 3 . The total number of states of the spin system is equal to the combinatorial set of all label choices including the null state ., The preservation of all neighborhood information at this step is particularly important for the analysis of data from larger proteins in which noise peaks and real peaks are closely interspersed ., The spin-system scoring step is used to integrate the spin system sub networks by assigning a score to each possible label that can be associated to a spin system ., This process makes use of empirical chemical shift probability density functions , calculated from combined BMRB ( chemical shift ) and PDB ( coordinate ) data from proteins of known structure , for each atom of every amino acid type in three label states: α-helix , β-strand , and neither helix nor strand ( other ) 13 ., The general form of the score is obtained by computing the probability of a chemical shift X having the label n ( residue number ) as described in Basic Algorithm: 8 . a ., This approach connects amino acid typing and secondary structure state determination through a conditional dependency model ., The successive application of weighted measures ( Basic Algorithm ) , leads to the definition of a complex network of relationships and weights among correlated sets of information at the global level ( Figure 3 ) ., This process establishes an initial system of neighborhoods ( Figure 2 ) ., Whenever an initial set of probabilities is unavailable , a uniform distribution is assumed as the starting state ., The challenge is to address the computationally demanding problem of deriving the backbone and side chain assignments from amino acid typing and other experimental data ( connectivity experiments ) according to the model described above ., Rather than modeling the assignment of labels to individual peaks , or assigning spin systems to a single amino acid , we generate triplet spin systems and label them to overlapping triplets of amino acids in the protein sequence ( Figure 5 ) ., The selection of tripeptides instead of single residues reduces the complexity of the graph by eliminating a substantial number of labeling choices; however , it may introduce additional noise to the network due to possible erroneous spin system assembly ., Given the trade-off between noise level and network complexity , we found that triplets yielded the optimum choice among other combinations of residues ., However , the resulting network of weights and relationships has a complex topology in which a large fraction of relationships ( edges ) arise entirely from noise in the data , and the resulting random field is not amenable to a straightforward implementation ., To overcome this problem , we determine , from spin system scoring and connectivity constraints , an initial topology and the sets of weights for the backbone ( Figure 6 and Basic Algorithm: 9 ) and side chain assignments ( Protocol S1 ) ., The topology ordinarily is dependent on the weights and a set of parameters ( thresholds ) ., These values typically are noisy and incomplete and are contaminated by false positives and false negatives ., Our goal is to evolve the initial state of the system toward an “optimally coherent” state without the need for any manual parameter settings by carefully managing the selection of network topology ., An initial topology for the network is determined by removing all edges with potential weights below a threshold value ., The threshold value is calculated ( Basic Algorithm: 10 ) automatically by approximating the level of success achievable by each threshold ( Figure S1 ) ., For a fixed set of edge values , this function is generally unimodal and defines the appropriate threshold for the starting state ., At each threshold , a variation of the belief propagation algorithm 42 operates on the dense multigraph to effectively prune many edges and to derive the posterior probabilities that define clusters ( or labels ) ., After each iteration step , the posterior probabilities of all label assignments are utilized to determine local topology modifications and new edge weights ., Secondary structure labels are dependent variables derived from prior chemical shift assignments ., Each chemical shift assignment has an associated probability , and we derive the probabilities for the assignment of secondary structure labels from a normalized and weighted sum of associated probabilities ., After computing the probability of each residue n to be in each of three conformational states ( s\u200a=\u200ahelix , strand , other ) by the method described in 13 for different assignment configurations , the overall secondary structure probability is calculated by Eq 5 ( Basic Algorithm ) ., Note that this step involves a shift in the point of view from chemical shift centric to residue centric ., Posterior probabilities derived in each iteration of the assignment process are used as local prior probabilities in the next round of assignment , provided that ( 1 ) the assignment has not been detected as an outlier , ( 2 ) the assignment of chemical shift is correlated with the assignment of secondary structure consistent with known empirical distributions , and ( 3 ) the assignment is consistent with established connectivity constraints ., If one or more of the above conditions are not met , the results are deemed inconsistent because the resulting probabilities appear as outliers of the marginals supported by the current graph topology ., This view is driven by the notion that the equilibrium of our fictitious system is the fixed point of the energy functional , with the factorization induced by our graph ., In order to reach the consistent state , scores are re-evaluated and a new local score is computed for the next iteration; a new topology is generated , and the computational steps are repeated ., The iteration process continues until a stationary or quasi-stationary state is reached , i . e . , when the topology of the network and the labeling probabilities do not vary significantly ., The iteration process leads to “self-correction” through appropriate adjustments to the topology of the underlying network in order to preserve maximum information ., PINE-NMR is designed to analyze peak lists derived from one or more of a large set of NMR experiments commonly used by protein NMR spectroscopists ., This set ( listed on the PINE-NMR website ) currently includes data types used for backbone and aliphatic side chain assignments ., ( PINE-NMR will be expanded in the future to handle aromatic side chain assignment . ), To test the software , we asked colleagues at the Center for Eukaryotic Structural Genomics ( CESG ) and the National Magnetic Resonance Facility at Madison ( NMRFAM ) to provide subsets of data from projects that had led to structure determinations with assigned chemical shifts deposited in the BMRB 49 ., We wanted the assignments to have been refined and vetted in light of a structure determination , because we took the BMRB deposited values to be “correct” ., In most cases , the input data supported the determination of both backbone and aliphatic side chain assignments ., In some cases , the input data supplied supported only the determination of backbone assignments ., The peak lists were provided by the persons submitting the data without any specification for the peak picking software , threshold , or other parameters ., Table 1 summarizes the PINE-NMR results for all datasets provided ., The input datasets are indicated along with the size of the protein ., A backbone or side chain assignment was scored as “correct” if the top ranked ( highest probability ) PINE-NMR assignment corresponded that in the BMRB deposition ., The assignment accuracy is given as the number of “correct” assignments divided by the total number of assignments supported in theory by the input data expressed as a percentage ., “The “correct” ( BMRB ) assignments had the benefit of additional information coming from NOESY data and filtering with respect to structure determination ., Also listed in Table 1 is the backbone “assignment coverage” achieved by PINE-NMR ( defined as the total number of correct backbone assignments in comparison to the total backbone assignments in the corresponding BMRB deposition expressed as a percentage ) ., The secondary structure accuracy reported in Table 1 compares the PINE-NMR result with the secondary structure of the deposited three-dimensional structure as determined by the DSSP software 50 ., It can be seen that the accuracy of the PINE-NMR results correlates with the data quality factor ., The outlier count is defined as the number of C′ , Cα , or Cβ atoms detected as possible outliers in the final assignment by the LACS method 16 ., In the majority of cases , the assignment accuracy was above 90% for backbone resonances and above 80% for aliphatic side chain resonances ., Two cases in Table 1 yielded assignment accuracies below 90% ., In the case of the 177-residue protein ( At5g01610 ) , the lower performance was due to the poor quality of data from a highly disordered region of the protein ., A human expert was unable to go beyond the PINE-NMR assignments , and additional data were required to complete the protein structure determination ., In the case of the 299-residue protein ( At3g16450 ) , its stereo array isotope labeling ( SAIL ) pattern 51 gave rise to chemical shift deviations that degraded expected matches ., In this case the performance of PINE-NMR could be improved by incorporating corrections for the deuterium isotope effects on the chemical shifts ., An illustration of the improvement achieved by combining information comes from comparing the assignment accuracy results from PINE with those from PISTACHIO 12 ( Table 1 ) ., PISTACHIO is an automated assignment tool developed earlier that does not make use of inferred secondary structure or outlier detection implemented in PINE-NMR ., The results from PINE-NMR also are superior to those achieved by iterative pipelining of the individual assignment ( PISTACHIO 12 ) , secondary structure determination ( PECAN 13 ) , and outlier detection ( LACS 16 ) steps ( results not shown ) ., The tests of PINE-NMR shown in Table 1 are highly stringent , in that minimal information is provided ., Separate tests ( results not shown ) demonstrate that the performance is improved if the input peak lists have been pre-filtered to correspond to spin systems ., The results of website users provide a separate measure of the performance of PINE-NMR ., Since July , 2006 , users have analyzed more than 1 , 300 sets of chemical shift data ., Without access to the final structures and chemical shift assignments for these proteins , these results could not be analyzed , as in Table 1 , with regard to correct assignments and secondary structure ., Instead , we used the results from Table 1 to estimate the empirical conditional probability of incorrect labeling in the user PINE-NMR output: P ( incorrect label| plabel\u200a=\u200ax ) ., Assignments with a probability higher than 0 . 95 generally were found to be correct ( Table 1 ) ., Using the data submitted to the PINE-NMR web site , we selected a representative sample of proteins with numbers of residues and data quality factors similar to those in Table 1 ., We then used the empirical estimate of accuracy to analyze the results from these proteins ( Table S1 ) ., The outcome was in substantial agreement ( in a statistical sense ) with the results shown

Introduction, Methods, Results, Discussion

The process of assigning a finite set of tags or labels to a collection of observations , subject to side conditions , is notable for its computational complexity ., This labeling paradigm is of theoretical and practical relevance to a wide range of biological applications , including the analysis of data from DNA microarrays , metabolomics experiments , and biomolecular nuclear magnetic resonance ( NMR ) spectroscopy ., We present a novel algorithm , called Probabilistic Interaction Network of Evidence ( PINE ) , that achieves robust , unsupervised probabilistic labeling of data ., The computational core of PINE uses estimates of evidence derived from empirical distributions of previously observed data , along with consistency measures , to drive a fictitious system M with Hamiltonian H to a quasi-stationary state that produces probabilistic label assignments for relevant subsets of the data ., We demonstrate the successful application of PINE to a key task in protein NMR spectroscopy: that of converting peak lists extracted from various NMR experiments into assignments associated with probabilities for their correctness ., This application , called PINE-NMR , is available from a freely accessible computer server ( http://pine . nmrfam . wisc . edu ) ., The PINE-NMR server accepts as input the sequence of the protein plus user-specified combinations of data corresponding to an extensive list of NMR experiments; it provides as output a probabilistic assignment of NMR signals ( chemical shifts ) to sequence-specific backbone and aliphatic side chain atoms plus a probabilistic determination of the protein secondary structure ., PINE-NMR can accommodate prior information about assignments or stable isotope labeling schemes ., As part of the analysis , PINE-NMR identifies , verifies , and rectifies problems related to chemical shift referencing or erroneous input data ., PINE-NMR achieves robust and consistent results that have been shown to be effective in subsequent steps of NMR structure determination .

What mathematicians call the “labeling problem” underlies difficulties in interpreting many classes of complex biological data ., To derive valid inferences from multiple , noisy datasets , one must consider all possible combinations of the data to find the solution that best matches the experimental evidence ., Exhaustive searches totally outstrip current computer resources , and , as a result , it has been necessary to resort to approximations such as branch and bound or Monte Carlo simulations , which have the disadvantages of being limited to use in separate steps of the analysis and not providing the final results in a probabilistic fashion that allows the quality of the answers to be evaluated ., The Probabilistic Interaction Network of Evidence ( PINE ) algorithm that we present here offers a general solution to this problem ., We have demonstrated the usefulness of the PINE approach by applying it to one of the major bottlenecks in NMR spectroscopy ., The PINE-NMR server takes as input the sequence of a protein and the peak lists from one or more multidimensional NMR experiments and provides as output a probabilistic assignment of the NMR signals to specific atoms in the proteins covalent structure and a self-consistent probabilistic analysis of the proteins secondary structure .

mathematics, biotechnology/protein chemistry and proteomics, biophysics/experimental biophysical methods, computational biology, chemical biology/protein chemistry and proteomics

journal.pcbi.1005603

Molecular determinants for the thermodynamic and functional divergence of uniporter GLUT1 and proton symporter XylE

The glucose transporter GLUT1 catalyzes facilitative diffusion of glucose into red blood cells1 and across the blood-brain barrier2 ., The bacterial homologues of GLUT1 are all proton symporters whereby the transmembrane proton gradient is employed to drive the uphill translocation of the substrate saccharides into the cell3 ., The distinct transport mechanisms are consistent with their working environment ., The glucose concentration in blood maintains at around 5 mM and the intake glucose is immediately metabolized to glucose-6-phosphate in the cytosol , thereby creating a constant transmembrane gradient of glucose4 ., A facilitative uniporter is thus sufficient to mediate the uptake of glucose ., In contrast , bacteria may have to hunt under stringent conditions ., A co-transport mechanism may ensure efficient uptake of the nutrient at low concentration in the environment ., Interestingly , GLUT1 and its bacterial homologues share considerable sequence similarities5 , raising the question of what is the determinant under the mechanistic divergence between the closely related proton symporter vs . uniporter ., The xylose:proton symporter XylE from E . coli is one of a number of rigorously characterized GLUT1 homologues ., In recent years , crystal structures of GLUTs6 , 7 , 8 and XylE were determined in multiple conformational states5 , 9 , 10 , in line with alternating access mechanism of membrane transporters11 ., Structure-guided mutational analysis identified Asp27 of XylE to be the protonation site for symport12 ., The D27N variant of XylE that was designed to mimic the neutral residue Asn29 at the corresponding position of GLUT1 , however , lost transport activity in in vivo experiments , despite fully active in in vitro counter-flow assays12 ., Thus , the behavioral difference between uniporters and symporters exemplified by GLUT1 and XylE cannot be accounted for simply from the perspective of protonation-site residues ., Instead , atomic-level description on the transitions between alternating-access states and quantitative evaluation on the thermodynamics of these processes are required to address these questions ., Molecular dynamics ( MD ) simulations , which resemble in silico single-molecule experiments at atomic resolution , emerge as a suitable tool for investigating conformational transitions of macromolecules13 ., In addition to the direct observation on large-scale conformational changes , thermodynamics and many physical properties could be rigorously evaluated to elucidate the internal causes of molecular behaviors14 , 15 , 16 , 17 ., In this work , we used MD simulations to study the alternating-access transitions of three apo systems: XylE with Asp27 protonated ( denoted as XylE_H ) , XylE with Asp27 deprotonated ( denoted as XylE_noH ) and GLUT1 ., Despite the lack of substrates , these simulations provided the quantitative details for the structural transitions from the inward-facing ( IF ) to outward-facing ( OF ) states , which are informative parts of the complete transport cycle ., From the free energy profiles calculated for the transitions , we not only revealed the coupling between Asp27 protonation/deprotonation and conformational transition of XylE , but also identified a remarkable thermodynamic difference between XylE and GLUT1 ., Subsequently , to further understand the mechanism of these thermodynamic observations , we developed Bayesian network ( BN ) models to analyze changes of residue interaction networks during conformational transitions ., Besides mechanistic illustration , these statistical models predicted a few residues essential for the appropriate conformational preference in XylE , which was then validated by experiments on the corresponding mutants ., More importantly , our results suggested that the thermodynamic divergence between XylE and GLUT1 arises from multiple residue substitutions accumulated during evolution ., Based on a group of residues inferred from our models , we successfully designed a uniporter-like XylE mutant , which was then confirmed by experimental validation ., Conclusively , our computation results provided insight into the mechanistic difference between proton symporters and uniporters in the absence of substrates ., GLUTs and XylE have a classical MFS fold of 12 transmembrane ( TM ) helices plus a unique intracellular helical ( ICH ) domain that comprises of 4 or 5 helices7 ., Structural comparison and accompanying biochemical characterizations of the sugar porter ( SP ) family members suggested that the two TM domains , the N-terminal domain ( NTD ) and C-terminal domain ( CTD ) , undergo concentric rotations relative to each other to accomplish transition between inward- and outward-facing conformations , resulting in the alternating access of the substrate binding site ( s ) from either side of the membrane18 , 19 ., To study the conformational transition , we started from the inward open GLUT1 and inward occluded XylE with Asp27 protonated/deprotonated denoted as XylE_H/XylE_noH hereafter ., The conventional MD ( cMD ) simulations ( Sim #1 , #2 and #3 in Fig 1B ) revealed little change with RMSDs of TM domains < 1 . 5 Å ( S1A Fig ) for all three systems ., Subsequently , accelerated MD ( aMD ) simulations were initiated to encourage conformational transitions without directional inclinations ( Sim #4 , #5 and #6 ) ., Conformational states captured by aMD simulations were illustrated on a 2D-map of Extracellular Gate distance and Intracellular Gate distance between NTD and CTD , which depict the extent of opening towards periplasm and cytoplasm respectively ( Fig 1A , see Free energy calculations in Methods section ) ., As shown in Fig 1C , GLUT1 rapidly left the initial state and underwent IF→OF transition ( also in S1B Fig ) , suggesting that the uniporter has a barrier-free energy landscape consistent with facilitative transport ., Unlike GLUT1 , aMD trajectories of XylE_H and XylE_noH were confined to inward- and outward-facing conformations respectively , without discernible IF↔OF transitions ( Fig 1C; S1C and S1D Fig ) ., Therefore , apo XylE might possess high-energy transition state ( s ) that would require substrate binding to stabilize ., We then performed extensive string method with swarms of trajectories ( SMwST ) simulations using conformations selected from the aMD and cMD trajectories for transition path identification ( Sim #7 , #8 and #9; see S2B Fig for convergence ) ., Further equilibrations of on-path images ( Sim #10 , #11 and #12 ) permitted path reconstruction in the space of gate distances and subsequent free energy calculations by bias-exchange umbrella sampling ( BEUS ) scheme ( S2A Fig ) ., In the BEUS simulations ( Sim #13 , #14 and #15 ) , each window/image represents a conformation of the IF→OF transition pathway ., The final free energies were evaluated ( see Free energy calculations in Methods section ) both in the space of gate distances ( S2C Fig ) and along window/image index ( Fig 1D ) ., Low RMSDs between structures in individual windows and the crystal structures of various states validate the efficacy of our path-finding protocol ( S1E Fig ) ., The three systems exhibit drastically different free energy profiles ( Fig 1D ) ., GLUT1 shows a broad energy well which is required for the fast turnover of passive uniporters ., The slight preference of OF over IF conformations agrees with the unidirectional conformational shift in the aMD simulation ( S1B Fig ) ., Notably , the GLUT1 profile exhibits no favorable energy wells at the window corresponding to the inward-facing crystal structure 4PYP ( S1F Fig ) ., The reason at least partially arises from the mutation and detergent introduced to the crystal structure ( see S1 Text ) ., In contrast , an energy barrier is present at the image index of 20 in both XylE systems ., According to our rough estimation , this barrier of ~5 kcal/mol can effectively forestall rapid IF↔OF transition ( Fig 1D ) and thereby suppressing proton leakage across the membrane in the absence of substrate ( see S1 Text for details ) ., The marked difference between XylE_H and GLUT1 at the energy barrier cannot be accounted for purely by the protonation-site residue ., In addition , comparison on free energy profiles between XylE_H and XylE_noH indicates that conformational preference of XylE could be altered thermodynamically by protonation/deprotonation ., Particularly , once the proton and substrate have been unloaded , inward-facing XylE would spontaneously restore the energetically more stable outward-facing state for another cycle of transport ., The systematic difference between XylE and GLUT1 may hinder accurate side-by-side comparison on their free energy profiles ., We thus developed an RMSD-scoring based algorithm to align BEUS windows for the three systems , which can effectively eliminate the cross-system window shift along the path ( see Window alignment in Methods section; S3 Fig ) ., From the post-alignment free energy profiles of the three systems , the IF and OF states as well as the transition state ( TS ) can be unanimously represented by windows renumbered as A3 , A16 and A12 respectively ( S2C Fig and S3C Fig ) ., Comparing aligned states of the three systems , the most noteworthy motion that can be visualized during transition is the domain rotation ( S1 Movie and S4B Fig ) ., Hence , we superimposed all structures upon their NTDs and investigated the global conformational changes using principal component analysis ( PCA ) on TM helices ( see Analysis techniques in Methods section ) ., The top 3 principal components explain ~85% of the structural variations in all systems and show strong inter-system correlations , which reflects consensus rocker-switch movements for both uniporters and symporters ( S4 Fig ) ., Local changes accompanying the conformational transition were characterized by monitoring the per-residue structural fluctuations among all aligned states ( S5 Fig ) ., As expected , NTD and CTD remain nearly rigid in majority of the TM helices ( RMSF < 1 Å ) ., In addition to the loop regions , TM7b exhibits flexibility within all three systems , and thus may be a local gating element coupled with the global conformational transition ., The TMs show different packing patterns in the extracellular and intracellular gates ., Unlike side-by-side helix anchoring in the former , TM5 and TM11 insert into the interfaces between TM8/10 and TM2/4 by partial twisting in the latter ( S4C Fig ) ., The relationship between global conformational transition and gating can be seen from the pore radius analysis ( S6C Fig ) ., Interestingly , we identified a unique gate around Tyr298 in XylE that confines the periplasmic entrance ( pointed by arrows in S6C Fig ) ., Although this gate becomes fully open in a transient state ( window A18 ) to allow substrate binding , its constriction tendency may cause considerably lower rate of substrate dissociation in XylE than in GLUT1 ., Apart from the global movement , TM7b bending contributes to the extracellular gating ( see the definition of kinking angle in S6A Fig ) ., In GLUT1 , the severe kinking of TM7b diminishes in the outward-facing state that exhibits striking structural resemblance to the outward-open GLUT3 ( PDB ID: 4ZWC ) ., Conversely , the helix remains bended in the outward-facing conformations of both XylE_H and XylE_noH , thus creating the unique gate ( S6 Fig ) ., A proline ( Pro301 ) residing in the middle of TM7b indicates why this helix is reluctant to adopt the straight conformation in XylE ., The above conformational changes cannot explain the functional and thermodynamic differences between uniporter GLUT1 and symporter XylE despite that they are consistent with the structural studies ., We then switched our focus to the interactions around the meaningful protonation-site residue , i . e . Asp27-Arg133 bonding in XylE and the corresponding Asn29-Arg126 interaction in GLUT1 , as reported5 , 10 ., Analysis on hydrogen bonds ( H-bonds ) reveals that side-chains of the Asn29-Arg126 pair in GLUT1 never form favorable interactions in any conformational states , whereas those of Asp27-Arg133 in XylE_H preserve moderate H-bonds in TS and OF states ( Fig 2; S1 Movie ) ., This observation and thermodynamic distinction between GLUT1 and XylE_H jointly negate the proposition that simple neutralization of Asp27 in XylE could eliminate their functional gap ., The Asp27-Arg133 bonding in XylE_noH system is well maintained in all conformations due to electrostatic attraction , which is consistent with structural observations10 ., In contrast , this interaction is greatly impaired in the inward-facing state of XylE_H ., Considering that XylE_H and XylE_noH only differs in the protonation state of Asp27 , changed bonding strength of Asp27-Arg133 could be the origin of their different thermodynamics ., To understand the molecular basis for the thermodynamic difference between XylE and GLUT1 observed in simulations , we made side-by-side comparison on the changes in residue interaction networks during their conformational transitions ., Conventional modeling strategies of residue interaction network usually construct undirected graphs with nodes representing residues or Cα atoms , and with edges reflecting correlated motions and/or physical contacts20 ., Although it has been reported that causality could be extracted from such networks21 , these mutual-information based approaches show limited predictive power on network tuning and had a poor performance in our case ., We designed a novel method of H-bond network modeling , named as the Interaction Regulation with Bayesian Networks ( IRBN ) , to infer the causal relationship between all interacting residue pairs and to search for factors responsible for the evolutionary divergence using the Bayesian network , which is widely used for causality inference in omics data analysis22 ., Instead of abstracting residues as nodes , inter-residue H-bonds were symbolized as basic units to construct network models , since they can bridge network intervention and pairwise interaction energies , the latter of which jointly correlate with free energy ( see Modeling of hydrogen bond networks in Methods section for details ) ., To reduce the network complexity , H-bonds formed by identical moieties of side-chain ( s ) ( SC ) and/or backbone ( s ) ( BB ) are treated as a whole ( Fig 3A; also see S7 Fig for comprehensive illustration of network learning and inference ) ., Intuitively , the aggregate number of H-bonds in a network ( defined as HB ) negatively correlates with free energy , and therefore the change of HB ( ΔHB ) could be used to assess perturbation in free energy ( ΔG ) for a specific conformational state induced by certain mutations ., To decipher the mutational effect on the free energy change during the transition between two conformation states ( ΔΔG ) , we extended the concept to ΔΔHB by mimicking a thermodynamic cycle ( Fig 3B ) ., To clarify distinct conformational preferences of XylE systems , we first constructed the network models for aligned IF , TS and OF conformations , and then disabled specific H-bonds of a polar/charged side-chain ( called mutation here ) in the models to calculate its ΔΔHB of OF→IF transition ., Positive ΔΔHBOF→IF signifies that the mutation introduces an additional bias to favor the inward-facing state ( in respect of HB ) , relative to the outward-facing state ., Fig 3C shows the ΔΔHBOF→IF values of the tested mutations in XylE systems ., The diagonal distribution demonstrated that the network interventions generally led to equivalent outcomes in XylE_H and XylE_noH systems ., The mutations that gave rise to substantial positive or negative ΔΔHBOF→IF values in both XylE systems were categorized as IF-favoring and OF-favoring respectively ( Fig 3D ) ., Mutation on either Asp27 or Arg133 favored inward-facing conformation , supporting our hypothesis that proton-coupled conformational preference originates from the modulation on Asp27-Arg133 bonding strength ., Mutations that weakened or destroyed the Asp27-Arg133 interaction , i . e . D27N and D27L , were tested by semi-quantitative PEGylation assays ( see PEGylation assay in Methods section; S8 Fig ) ., As shown in S1 Text , results of PEGylation for certain mutation plus L65C or V412C should be interpreted by comparison with L65C or V412C , respectively ., Thus , we can conclude that the population of outward-facing states dramatically declined in D27N and completely diminished in D27L , comparing to WT XylE ( S8C Fig ) ., The results support that Asp27 protonation modulates conformational preference in XylE through adjusting the local H-bond network ., Other residues whose mutations showed significant conformational preference were mainly involved in inter-domain interactions , including residues located at the NTD-CTD interface and conserved cytoplasmic motifs in SP family ( Fig 3D ) ., PEGylation assays showed that notable changes could only be detected in the mutants with markedly perturbed salt bridges , such as D337L , D27L and K305M , possibly because salt bridges contribute more to free energy than regular H-bonds ., Similar to D27L , the D337L mutant strongly prefers inward-facing state ( S8C Fig ) ., Despite that mutation on Lys305 ( K305M ) was expected to favor outward-facing conformations , its impact is weaker than D27L and D337L , considering the smaller magnitude of its ΔΔHBOF→IF value ( less than cutoff , see Fig 3C and S8C Fig ) ., Since Lys305 is the only salt-bridge forming residue predicted to favor outward-facing state , we constructed double mutants combining K305M and mutations in OF-favoring region of Fig 3C including Y298F and Y179F ., Consistent with model prediction , in comparison to the WT XylE , the inward-facing conformations almost disappear in the tested double mutants ( S8C Fig ) ., We thus confirm the predictive power of the newly developed methodology ., Likewise , the same analysis on GLUT1 predicted multiple potential mutations for stabilizing typical conformations , which awaits further experimental validation ( S8B Fig ) ., To explore detailed mechanisms for the mutational perturbation on HB values , we evaluated the structure of Bayesian network models by model averaging ( S7C Fig ) and presented the individual nodes that were intensely perturbed upon network tuning ( see Modeling of hydrogen bond networks in Methods section ) ., Considering the coincidence of Asp27-Arg133 bonding and free energy discrepancy between XylE_H and XylE_noH ( see Fig 1D and Fig 2B ) , we disrupted the Asp27-Arg133 side-chain H-bonds in network models but retained other interactions concerning Asp27/Arg133 ( e . g . , Asp27-Glu206 ) ( Fig 4A ) ., Following this protocol , we investigated the change in the architecture of H-bond networks upon Asp27-Arg133 H-bond dismissal in XylE systems ., When forcing the number of Asp27SC-Arg133SC H-bonds to zero for Bayesian network inference , the overall HB value of IF state is considerably less weakened than those of TS and OF states in both systems , suggesting that breaking this interaction triggers inward-facing preference ( Fig 4B ) ., Specifically , in the XylE_H system ( Fig 4C ) , Asp27SC-Arg133SC disruption introduces no changes in IF state , whereas numerous inter-helix H-bonds are weakened in TS and OF states ., Moreover , the upheaval of network architecture in diverse states revealed drastic variation of interaction patterns ., In the XylE_noH system ( Fig 4D ) , unlike the TS and OF states , H-bond loss regarding Asp27 in IF state ( i . e . Asp27SC-Arg133SC and Asp27SC-Glu206SC ) can be partially compensated by multiple connected interactions ( i . e . Ser102SC-Arg133SC and several TM4 backbone H-bonds ) ., We speculate that the sacrifice of these compensated H-bonds upon Asp27SC-Arg133SC fortification may be the cause of unfavorable inward-facing conformation in deprotonated XylE ., Complete removal of the H-bonding capability of Asp27 triggers more intricate network perturbations , yet generates a comparable ΔHB pattern destabilizing TS and OF states more than the IF state ( S9 Fig ) ., However , similar treatment with Asn29 side-chain in GLUT1 does not induce strong IF or OF preference ( S9D Fig ) , possibly because of the dramatic rearrangement of local H-bond network as exemplified by the accumulation of hydrophobic residues surrounding Asn29 in GLUT1 ( S9A Fig ) ., Therefore , multi-fold evolutionary events may have occurred to modulate the local environment surrounding Asp27 in XylE and Asn29 in GLUT1 in divergent directions ., As shown in the free energy landscapes , the most crucial discrepancy between XylE and GLUT1 is the presence/absence of energy barrier at transition state ., To elucidate its molecular basis , we sought for side-chain mutations in XylE_H and GLUT1 systems that would alter the energetics towards each other based on Bayesian network inference ( Fig 5A ) ., Considering the potential negative correlation between free energy and total amount of H-bonds ( HB ) , a GLUT1 mutation that devastates H-bonds at transition state more than IF/OF states is likely to convert its free energy profile towards XylE_H pattern , and vice versa ., Possible determinants for evolutionary divergence can be predicted from the values of ΔΔHBIF→TS and ΔΔHBOF→TS , thusly ., Since it was impractical to consider all possible combinations , we only focused on screening single side-chain mutations of GLUT1 and XylE ., The mutations that would generate a barrier in GLUT1 ( located in the blue region of Fig 5B ) and vice versa in XylE_H ( located in the red region of Fig 5B ) , were itemized in Fig 5C ., After disposal of unaligned , conserved and binding-site residues in both transporters , a total of 7 residues were identified in GLUT1 to account for its thermodynamic divergence from XylE ., These residues are H-bonding donors/acceptors in GLUT1 but are replaced by nonpolar residues with comparable volume in XylE ( Fig 5C ) ., To verify their joint performance , we mimicked a combination of 7 point mutations to replace the XylE residues with the corresponding ones in GLUT1 ( T45V , T60A , D236L , S294A , T295P , S313V and Y424M ) in the Bayesian network models of GLUT1 and re-evaluated the HB loss in various states ( Fig 5D ) ., Just as expected , the HB value of the mutant decreases dramatically at TS state ( by > 3 ) , thus supposedly converting the flat energy landscape of GLUT1 into a XylE_H-like fashion and prohibiting rapid IF↔OF transition ., Unfortunately , we could not obtain well-behaved protein of these GLUT1 mutants required for biochemistry characterization ., As a compensation , we tested the effect of these residues in the reverse direction , trying to convert XylE into a uniporter ., Besides the 7 candidate residues , D27N was also introduced into XylE , and the mutants were evaluated by in vivo transport assays ( Fig 5E ) ., The comparable transport activity between D27N variant with negative control indicates that D27N mutation alone is insufficient for the symporter-to-uniporter conversion ( see S11A Fig and S1 Text ) ., Similarly , incorporating a single mutation ( V43T , A52T , A300S , P301T , L248D , V321T , or M428Y ) with D27N presented little to no activity increase ., However , the combination of all mutations led to a significantly higher transport efficiency and thus indeed converted a symporter to a functional uniporter ., In vitro counterflow assays also confirm the functionality of this 7-mutation variant comparing to uniporters GLUT1 and GLUT3 ( S11B Fig ) ., Notably , the identified residues are not confined to a sub-region of the structure ( Fig 5D ) , implying that elaborate adjustment of interactions rather than simply altering protonation-site residues may be essential for driving uniporters and symporters to divergent directions ., We hereby present multiplex analyses for the sugar porter family members XylE and GLUT1 using MD simulations and a novel approach based on Bayesian networks for residue interaction analysis ., The free energy calculations on the 3 systems discriminate XylE from GLUT1 , and highlight the protonation state of Asp27 as the molecular basis for conformational preference alteration in XylE ., Inspection of IF↔OF transitions reveals numerous details that are consistent with structural and biochemical studies , supporting the reliability of aMD+SMwST path-finding scheme ., The newly developed modeling framework IRBN plays a pivotal role in mechanism illustration ., For instance , from the network models , we can predict that H-bond loss following the disruption of Asp27-Arg133 interaction in the inward-facing state of XylE_noH will be compensated by three contiguous interactions ( Fig 4C ) ., By artificially manipulating residue interactions , IRBN helped disclose molecular determinants for the uniporter/symporter divergence , hence implying that complex variations instead of mere Asp↔Asn substitution are required for the functional interconversion ., As for GLUT1 , we could also explain the mechanism of some disease-related mutations ., As shown in S10A–S10F Fig , these mutations unanimously perturb the energetic balance among states by disrupting local H-bond network , and consequently impair fast turnover required for facilitators ., Moreover , our approach of combining MD simulations and IRBN modeling could be extended to the fields of protein engineering and drug development , and could facilitate rational design as well as allosteric regulation studies for pharmaceutically important targets ., Under the guidance of computer simulation and statistical modeling , we successfully transformed a symporter ( XylE ) into a uniporter by neutralizing protonation site residue and reducing the energy barrier ., In contrast , converting GLUT1 to a proton symporter would presumably need more intricate changes to meet following requirements: ( 1 ) ability to load and unload proton , ( 2 ) no proton leak in the absence of substrate , and ( 3 ) no substrate leak without change in the protonation state ., Here , we provided several necessary tactics for the uniporter-to-symporter design , which were generalized from our calculations on thermodynamics: ( 1 ) possess a titratable residue as protonation site , ( 2 ) create an energy barrier in apo state , and ( 3 ) destabilize inward-facing state in deprotonated state ., Notwithstanding these advances , it is noteworthy that relative free energies between IF and OF states in XylE_H and GLUT1 systems slightly disagree with experimental observations23 for two possible reasons: ( 1 ) the applied force field did not consider polarizable dipole-dipole interactions that were supposed to fasten the extracellular gate7 , and ( 2 ) boost energies in aMD simulations partially destroyed the integrity of ICH domain in the outward-facing conformations , which should be intact as exemplified by outward-facing GLUT3 structures ( PDB ID: 4ZW9 , 4ZWB , and 4ZWC ) 6 ., In summary , we thoroughly investigated crucial reactions of the transport cycles emphasized in Fig 5F ., Comparison of three apo systems provides detailed understanding of transporter mechanisms ., Since sugar porters may recognize both α- and β-anomers as substrates , the uncharacterized processes in Fig 5F still await extensive computational and biophysical research ., Using the plugins in VMD24 , we established three systems ,, i . e ., Asp27 protonated XylE ( XylE_H ) , Asp27 deprotonated XylE ( XylE_noH ) and GLUT1 ., The structure of inward-open GLUT1 mutant ( N45T & E329Q , PDB ID: 4PYP ) were preprocessed for MD simulations with 3 modifications: ( 1 ) Gln329 was mutated back to Glu as in WT GLUT1 , ( 2 ) the detergent β-NG whose head group occupies the binding pocket was removed , and ( 3 ) the missing ICH5 ( residue 459 to 468 ) was constructed as α-helix referring to template structures of the outward-facing XylE ( PDB ID: 4GBY ) and GLUT3 ( PDB ID: 4ZWC and 4ZW9 ) using Modeller 9v1225 ., For XylE , we selected the inward-occluded structure ( PDB ID: 4JA3 ) , and modeled all missing loops using the outward-occluded conformation ( PDB ID: 4GBY ) as the template ., PROPKA 3 . 126 , 27 was used to determine the protonation states of titratable residues other than Asp27 in XylE at pH 7 . 0 ., In specific , Glu206 was neutralized in both XylE systems ., XylE was inserted into a palmitoyl-oleoyl-phosphatidyl-ethanolamine ( POPE ) bilayer , given the fact that the majority of E . coli ., lipids ( ~75% ) belong to the PE class28 , ., To mimic the physiological condition for GLUT1 , a palmitoyl-oleoyl-phosphatidyl-choline ( POPC ) membrane was adopted since it has been reported to restore transport activity of purified GLUT129 ., After solvation and neutralization in 150 mM NaCl , the total number of atoms reached ~86 , 000 for each system ., We generated input files and performed simulations using MD simulation suites AMBER1230 and AMBER1431 ., The transporters were parameterized by ff12SB force field , and were surrounded by LIPID11 phospholipids32 and TIP3P water molecules33 ., With the protein and ligand fixed , the systems first underwent a 5000-step minimization ., Then , a 1-ns melting of lipid tails was simulated in an NVT ensemble at 310 K with the rest of the system constrained with a large force constant k = 100 kcal/mol/Å2 ., Afterwards , one heating procedure ( k = 10 kcal/mol/Å2 for protein ) was carried out from 0 K to 310 K under constant volume condition for 1 ns ., Next , the value of k was set to 1 kcal/mol/Å2 for another 1-ns pre-equilibration ., To further naturalize the lipid bilayer , two 5-ns runs in the NPγT ensemble ( 1 atm of pressure ) were performed with k = 0 . 1 kcal/mol/Å2 on Cα atoms and no constraint at all , sequentially ., We deliberately selected surface tension γ for bilayers ( γ = 17 dyn/cm for POPC and γ = 26 dyn/cm for POPE membrane ) suggested by the reported tests of LIPID11 force field32 ., Under periodic boundary conditions ( PBC ) , all pre-equilibrations were performed with the time step of 1 fs and the van der Waals cutoff of 10 Å , using the Particle Mesh Ewald ( PME ) method to estimate electrostatics34 ., Initially for each system , we performed an NPγT cMD simulation for 100 ns ( Sim #1 , #2 and #3 ) using the time step of 2 fs and with the SHAKE algorithm35 applied ., Tiny fluctuations of the system volume indicated that membrane and solvent molecules were well equilibrated , and therefore we fixed volume and used the GPU implementation of PMEMD36 for the subsequent simulations ., It is usually unrealistic to sample the large-scale conformational change of transporters by cMD simulations , because of the long autocorrelation time ranging from micro- to milliseconds or even longer ., Therefore , we conducted aMD simulations ( Sim #4 , #5 and #6 ) to sample the conformational transition , based on a reported study of GPCR37 ., Preserving the shape of energy surface , aMD adds a boost potential ΔV, ( r ) to adjust total potential and/or dihedral potential14:, V, ( r ) *=V, ( r ) +ΔV, ( r ) ,, ΔV, ( r ) = ( Ep−V, ( r ) ) 2αp+Ep−V, ( r ) + ( Ed−Vd, ( r ) ) 2αd+Ed−Vd, ( r ) ., Here Ep and Ed denote the reference values for the total and dihedral potentials respectively , while V, ( r ) and Vd, ( r ) denote the total and dihedral potentials calculated for the current state of the system ., The boost energy that is tuned by the parameters αp and αd is applied only in the situation of V < E . We set Ep and Ed as the average potentials of the second halves of the 100-ns cMD trajectories , and expressed the parameters via:, Ed=Vd⋅ ( 1+λd ) ,, αd=λd5⋅Vd ,, Ep=V+λp⋅Natom ,, αp=λp⋅Natom ., After testing some combinations of parameters , we found that enhanced sampling could be visualized within 100 ns for λd = 0 . 3 and λp = 0 . 2 , without perturbing the system stability ( < 5% loss of helical contents ) ., Subsequently , we extended the aMD simulations to 500 ns for all systems ., Flattening energy barriers on the path , aMD allowed the protein structure to evolve faster and presented an overview of

Introduction, Results, Discussion, Materials and methods

GLUT1 facilitates the down-gradient translocation of D-glucose across cell membrane in mammals ., XylE , an Escherichia coli homolog of GLUT1 , utilizes proton gradient as an energy source to drive uphill D-xylose transport ., Previous studies of XylE and GLUT1 suggest that the variation between an acidic residue ( Asp27 in XylE ) and a neutral one ( Asn29 in GLUT1 ) is a key element for their mechanistic divergence ., In this work , we combined computational and biochemical approaches to investigate the mechanism of proton coupling by XylE and the functional divergence between GLUT1 and XylE ., Using molecular dynamics simulations , we evaluated the free energy profiles of the transition between inward- and outward-facing conformations for the apo proteins ., Our results revealed the correlation between the protonation state and conformational preference in XylE , which is supported by the crystal structures ., In addition , our simulations suggested a thermodynamic difference between XylE and GLUT1 that cannot be explained by the single residue variation at the protonation site ., To understand the molecular basis , we applied Bayesian network models to analyze the alteration in the architecture of the hydrogen bond networks during conformational transition ., The models and subsequent experimental validation suggest that multiple residue substitutions are required to produce the thermodynamic and functional distinction between XylE and GLUT1 ., Despite the lack of simulation studies with substrates , these computational and biochemical characterizations provide unprecedented insight into the mechanistic difference between proton symporters and uniporters .

We seek to address one intriguing question , the mechanistic distinction between active proton-coupled symporters and passive uniporters that are related in evolution ., Proton-coupled symporters harness the transmembrane proton gradient to drive the substrate transport , while uniporters can only facilitate the passive substrate translocation ., In this work , we focus on two sugar transporters GLUT1 and XylE , which belong to symporters and uniporters respectively but have high sequence similarity ., We first applied molecular dynamics simulations to characterize the thermodynamic behaviors of apo GLUT1 and XylE , which are supposed to provide prominent details of mechanisms ., From the identified difference in thermodynamics , we concluded that neutralizing protonation site in XylE is insufficient for its conversion to GLUT1 analog ., To pinpoint extra elements contributing to their evolutionary divergence , we developed a novel network modeling scheme based on Bayesian network which shows impressive predictive power on residue mutations ., Our models suggested the detailed mechanism of proton coupling in XylE and molecular basis of symporter/uniporter discrepancy ., Furthermore , the modeling scheme could help to guide the design of biomolecules for desired functions .

protons, crystal structure, condensed matter physics, network analysis, crystallography, thermodynamics, computer and information sciences, solid state physics, proteins, pegylation, biophysics, nucleons, free energy, physics, biochemistry, biochemical simulations, nuclear physics, post-translational modification, biology and life sciences, physical sciences, computational biology, biophysical simulations

journal.pcbi.1005485

Modeling disordered protein interactions from biophysical principles

Intrinsically disordered proteins ( IDP ) , which have evolved to not adopt a stable structure under physiological conditions , are a departure from the traditional paradigm of structured proteins 1 ., After initial recognition of their critical biological functions in the 1990s 1 , IDPs quickly gained attention as they were found to be abundant in genomes across all three kingdoms 2 ., IDPs are known to be involved in many molecular recognition events ., Particularly , it is estimated that 15–45% of protein-protein interactions ( PPIs ) are formed with IDPs 3 ., A well-known example is the p53 tumor suppressor , which contains disordered regions that interact with dozens of partner proteins 4 ., Due to the abundance and characteristic features of IDPs in PPI networks , including many critical signaling pathways , fully understanding the molecular mechanisms of PPI networks requires consideration of the role of interactions with IDPs ., The binding mechanism of an IDP to a structured target protein , i . e . a disordered PPI , has drawn much interest in the context of binding rate constants , because disordered PPIs achieve high specificity and high dissociation rate constant simultaneously , which is an ideal characteristic for signaling pathways but difficult to realize with interactions of structured proteins 5 ., It is generally accepted that binding precedes global folding of the IDP , although secondary structures in local regions may form before the interaction ., In the model called the dock-and-coalesce 5 , a small segment of the IDP , which may be folded into secondary structure prior to binding , forms the initial contact with the ordered partner , followed by coalescence of the rest of the IDP into the bound conformation ., This mechanism imparts both thermodynamic and kinetic advantages ., Forming a binding interface out of segments leads to a large interface with fewer amino acids than a structured protein 2 , 6 and the binding affinity is accumulated from the affinities of each segment 5 ., This allows IDPs to have high binding specificity , but the loss of entropy upon binding imparted by the flexibility makes the interaction reversible 7 ., From a kinetic perspective , sequential binding of individual segments will have a much higher rate constant than a hypothetical situation in which a pre-organized IDP simultaneously makes all contacts with the ordered protein 5 ., A computational method based on the dock-and-coalesce model was successful in predicting the binding rate constants of disordered PPIs 8 ., Experimental structure determination of disordered PPIs using techniques such as X-ray crystallography and nuclear magnetic resonance ( NMR ) is challenging due to the flexible nature of IDPs and their tendency to form weak , transient interactions 9 ., Indeed , not all IDPs form a single , stable structure when bound ., Examples of these so-called “fuzzy” complexes are cataloged in FuzDB 10 , 11 ., Along a similar line , pE-DB contains ensembles of conformations that can be adopted by an IDP 12 ., Nevertheless , many proteins annotated as disordered in DisProt 13 do adopt a bound structure that can be experimentally determined ., For PPIs of structured proteins , experimental structure methods can often be complemented by computational modeling of protein complexes ( docking ) 14 ., However , current rigid-body and flexible docking methods ( which allow small conformational changes at the docking interface ) are not able to model disordered PPI prediction , because the required rigid structures are not available for IDPs ., Among existing protein modeling techniques , peptide-protein docking methods would be the most similar to disordered PPI prediction ., Approaches to peptide-protein complex modeling include template-based modeling ( TBM ) 15 , 16 , molecular dynamics ( MD ) 17–19 , small molecule docking 20 , 21 , protein-protein docking with flexibility 22–26 , and coarse-grained docking 27 ., The characteristics of the docking and MD methods are compared in Table 1 ., Several of the methods require knowledge of the binding site as input ., Information about the binding site can be obtained experimentally or by using computational prediction of peptide binding sites 28–30 or protein binding sites 31 , 32 ., More fundamentally , existing methods were developed and tested for binding short peptides of 2–16 residues , which is far shorter than the 10–70 residue IDPs that participate in disordered PPIs 2 , although some programs are able to accept peptides up to 30 residues in their web servers ., To predict the tertiary structure of a disordered PPI , a method must solve two interdependent problems: the tertiary structure of the input sequence of the disordered protein and its binding location on the receptor protein ., This is a difficult task as the conformational space to be explored for an IDP is enormous and grows with its length ., Currently , no existing methods can dock a long disordered protein to its receptor protein ., A totally new approach is required for predicting the structure of a disordered PPI involving commonly observed long IDPs ., In this work , we describe the development of a novel computational method named IDP-LZerD , which is able to model for the first time the docked structure of long IDPs ( 15–69 amino acids ) ., IDP-LZerD applies the biophysical principles of the dock-and-coalesce mechanism of IDP binding to model the structures of long IDPs ., In the “dock” phase , small segments of the IDP are modeled in various conformations and docked globally to the ordered protein ., Modeling and docking small segments is not only faster and easier but also consistent with the biophysical mechanism of small segments of the IDP binding sequentially ., In the “coalesce” phase , the docked segments of neighboring regions of the IDP are found and combined into a complete structure of the disordered PPI ., We found that correct bound conformations of the IDP were selected using scores evaluating docking with the receptor , which corresponds to the biophysical model that the conformation of an IDP is stabilized and determined by contacts with its receptor ., In addition , the combination of the docking scores of multiple segments is analogous to the accumulation of the binding affinities of multiple segments 5 ., Overall , we show that IDP-LZerD is able to yield docking models of a practical quality in a number of bound and unbound structures of PPIs involving long IDPs ., Secondary structure was predicted for each IDP using JPRED 44 , Porter 45 , SSPro 46 , and PSIPRED 47 ., The secondary structure predictions were reasonably accurate ( Table 4 ) ., If the predictions are considered correct when any of the four methods predicts the correct secondary structure , the accuracy is 86% ., For 57% of residues , all four methods predicted the correct secondary structure ., Even in the minority of cases where none of the methods predicted the correct secondary structure , fragments of all three secondary structure classes were created ( described below in Methods ) ., The full sequence of a target disordered protein was divided into 9-residue windows with a 3-residue overlap ., Fragment structures of each window were predicted using Rosetta Fragment Picker ( RFP ) 49 , which predicts structures based on the sequence profile 50 and predicted secondary structure 44–47 ., RFP was configured to output 30 fragments for a window ., Increasing the number of fragments chosen did not yield structures of a substantially lower root mean square deviation ( RMSD ) to the native structure ( S1 Fig ) ., Fragment structure was predicted reasonably accurately: on average the largest backbone RMSD of 30 conformations for a window was 1 . 8 Å for the training set , 1 . 6 Å for the test set , and 1 . 8 Å overall ( S2 Table ) ., For a sequence window , each of the 30 fragment structures was docked with the receptor protein using LZerD 35–37 ., LZerD is a shape-based , rigid-body docking method with the advantage of a soft representation of the surface shape of a protein that accounts for some conformational change upon binding ., Docked fragment poses were clustered and the top 4 , 500 cluster centers were selected ( see Methods ) ., Ranking was performed using the sum of the Z-scores of two scoring functions , DFIRE 51 and ITScorePro 52 , named DI score ., DI score was shown to perform better in docking pose selection than the individual scores ( S3 Table ) ., The docking accuracy of fragments is summarized in the “All docked” columns in S2 Table ., For bound cases , on average the worst ( largest ) of the minimum L-RMSD from all the windows in a target was 3 . 7 Å and 4 . 1 Å for the training and the testing set , respectively ., For unbound cases , the values were slightly worse , 4 . 4 Å and 4 . 3 Å for the training and the testing set , respectively ., Fragment structure and docking accuracy was further tested on an additional independent test set of 11 cases of 9-residue IDP complexes found in the database of eukaryotic linear motifs ( ELMs ) 53 ( Table 5 ) ., The results are shown in Table 6 ., The average fragment RMSD is 1 . 4 Å and the average minimum docked RMSD is 3 . 2 Å for both bound and unbound cases ( Table 6 ) , which are better than the results shown in S2 Table ., Selection of docked fragments was successful for most of the training set complexes , with an average RMSD of 5 . 4 Å for bound and 6 . 5 Å for unbound ( “Selected docked” columns in S2 Table ) ., On the testing set , the results are similar , 5 . 7 Å and 6 . 3 Å for bound and unbound cases ., Exceptions included 2clt and 1bk5 , where poor selection of docked fragments prevented successful modeling in the subsequent steps ., On the additional ELM-derived dataset , results were 4 . 9 Å for bound and 4 . 7 Å for unbound ( Table 6 ) , which are again comparable to the results on the testing and training datasets ., Interestingly , as shown in Fig 2 , evaluating docking fit with DI score often identified fragments of a low RMSD ., To understand the general trend , for each sequence window we compared the fragment RMSD distributions of the 30 fragment structures from RFP and the top 30 docked fragments by DI score ., Out of 144 windows from the 28 cases in the training set , for 83 ( 57 . 6% ) windows the top 30 by DI score are either better ( p <0 . 05 by the Mann-Whitney U test ) or contained five or more fragments with an RMSD better than 3 . 5 Å ( considered because there were cases where all 30 fragments from RFP were below 3 . 5 Å RMSD and no further improvement is possible by the DI score choice ) ., This indicates that the DI score is detecting the increased binding affinity of the correct conformation when bound in the correct location , analogous to induced fit upon binding ., Docked fragments from each window were combined to form full-length IDP complexes , referred to as paths ., First , we performed a pre-filtering of docked fragment pairs , which removes physically improbable pairs by considering mutual distances and angles; then , paths were assembled using an extend-and-cluster strategy ( see Methods ) ., This procedure effectively reduced the search space from as many as 1041 to the order of 105 paths regardless of the length of the IDP ( S2 Fig ) ., Overall , the combination process successfully produced low RMSD paths ., Out of the fourteen IDPs in the training set , for eleven bound and eight unbound receptors , paths with a 6 . 0 Å or lower RMSD were constructed ( “Clustered paths” in S2 Table ) ., Results were slightly worse for the testing set , an RMSD of below 6 Å was obtained for three bound and three unbound cases out of the eight IDPs ., For a complex , up to 1000 paths were chosen for further refinement ., Paths were scored using a linear combination of four terms ( Path Score ) : the energy score , representing the docking scores of fragments across all windows; the overlap score , evaluating how well the neighboring docked fragments fit into a continuous path; the cluster size , accounting for the consensus of docking poses; and the receptor score , which measures docking site consensus ., Path Score selected more hits than any of the individual score components ( S4 Table ) ., On average , the minimum RMSD of selected paths was 6 . 7 Å for bound and 8 . 0 Å for unbound in the training set and 7 . 5 Å and 8 . 2 Å for bound and unbound in the testing set ( S2 Table ) ., As in the situation in the docked fragment selection ( Fig 2 ) , it was observed that Path Score selected many models with IDPs of correct conformation ( RMSD under 6 . 0 Å; Fig 3 ) ., Out of the fourteen pairs of targets in the training set , in ten/eleven cases for bound/unbound at least one of the top 10 models by Path Score has a correct IDP conformation ., For the testing set , in four out of eight cases for both bound and unbound Path Score selected a correct IDP conformation within the top 10 ., These are again interesting results because Path Score mainly evaluates the binding affinity of a target IDP and its receptor , but also identifies IDPs of the correct conformation ., Thus , in accordance with the biophysical mechanism , the binding affinity of the IDP is accumulated from the binding affinities of the individual segments and the conformation of IDPs is determined by binding ., Selected paths underwent structure refinement using constrained molecular dynamics , which connects neighboring fragments in a path and relaxes the overall IDP structure ., An initial structure of a path was created by averaging the positions of the overlapping atoms ( Fig 4A , purple ) ., Multiple rounds of minimization were performed using tapering harmonic restraints to prevent excessive movement of fragments ., Refinement improved the protein-like nature of the combined fragments in a path ., Before refinement , only 50 . 4 ( 48 . 2 ) % of ligand Cα-Cα distances were between 3 . 75 and 4 . 0 Å in the training ( testing ) set , which was improved to 92 . 3 ( 95 . 8 ) % by the refinement ( S3A Fig ) with a small cost of deterioration of ligand RMSD ( L-RMSD ) for about half of the cases ( S3B Fig ) ., In parentheses , results for the testing set are shown ., Refinement improved both L-RMSD and rank for some models , including the first hit for Bcl2-like protein 1 ( Bcl2-L-1 ) and its antagonist ( BAD; PDB ID 2bzw; Fig 4A ) ., Originally , the path was ranked at 14 with a L-RMSD of 4 . 40 Å , which improved to rank 1 with L-RMSD 3 . 75 Å by the refinement ., Finally , refined models were re-ranked and selected using a composite score of DFIRE 51 , ITScorePro 52 , a molecular mechanics score 54 , and GOAP 55 ( Model Score ) ., Model Score selected hits at a higher rank than the single scores ( S5 Table ) ., Model Score has moderate overall correlation to L-RMSD but often selected acceptable models with low scores ( Fig 5 , left panel ) and successfully identified hits in many cases as we discuss in the next section ., RMSD of IDPs only and L-RMSD of docked models only correlate for models with an L-RMSD less than 10 Å ( Fig 5 , right panel ) ., Tables 7 and 8 summarize prediction results on the training and testing sets , respectively , listing the rank of the first acceptable model ( RFH ) ( the criteria for an acceptable model are shown in S1 Table ) and fnat ., On the training set ( Table 7 ) , IDP-LZerD produced at least one hit within the top 1000 models for thirteen bound and eleven unbound targets , and Model Score ranked hits within the top 10 for ten bound and five unbound cases ., Notably , the rank 1 model was a hit for four complexes ( three bound , one unbound ) ., There was only one complex where no hits were produced for both bound and unbound ( 2c1t/1bk5 ) ., On the testing set ( Table 8 ) , IDP-LZerD produced at least one hit within 1000 models for almost all of the targets: all eight bound and seven unbound targets , and one top 1 hit for both bound and unbound ., These fractions of top 1000 hits are higher than on the training set ., Hits were ranked in the top 10 for two bound and three unbound cases ., The fraction of top 10 hits ( 2/8 , 25% , for bound cases ) is lower than for the result observed on the training set ( 10/14 , 71% ) , while higher for unbound cases ( 3/8 , 37 . 5% ) than the training set ( 5/14 , 35 . 7% ) ., Interestingly , for most of the cases in both training and testing set results , the acceptable models have a high fnat , much higher than the 0 . 1 minimum for an acceptable model defined by CAPRI ( S1 Table ) ., A high fnat indicates that binding positions of IDPs are well reproduced in the models ., We also evaluated predictions in terms of the fraction of correctly placed ligand residues of the top 10 models ( BF10 ) ., Unsurprisingly , the fraction is high for cases with hits ranked in the top 10 ., What is more interesting is that there are cases where targets that do not have any hits within the top 10 nevertheless have substantial BF10 , which indicate largely correct models are ranked high ., Such targets include 1wkw , 1l8c , 2o8g , and 1l2w from the bound targets and 1ipb , 4i9o , 1khx , and 1u2n from the unbound targets ., Fig 6 shows examples of four bound and four unbound complexes with acceptable or better top 10 hits ., The four bound cases shown , 1ycr , 2cpk , 3owt , and 1xtg , include two medium quality hits , with RMSD at the interface ( I-RMSD ) below 2 . 0 Å ( 1ycr and 2cpk ) , and the IDPs range in length from 15 to 59 amino acids ., The four unbound cases , 4ah2 , 1ijj , 1l3e , and 1jya , have IDPs between 20 and 69 amino acids ., In all these examples , binding sites of the receptor proteins were accurately identified and overall docking structures were well predicted; often , even the pitch of the helices was reproduced ., These examples demonstrate that IDP-LZerD can successfully select and combine docked fragments to produce accurate top 10 models for IDPs , even for cases with well over 30 amino acids ., We also tested if binding residue predictions of receptor proteins is useful to improve model selection ( Table 7 ) ., We used BindML 56 , which predicts binding site residues from their mutation patterns ., Models were first filtered by the agreement of binding residues to the BindML prediction ( S4 Fig ) ; then , the selected models were ranked by Model Score ., Using BindML prediction ( Table 7; RFH-B ) did not make a large difference but slightly improved the model selection performance for 10 cases without worsening any cases ., In this section we evaluated the impact of secondary structure prediction on the quality of final models in two ways ., First , in S5 Fig we examined how the accuracy of the secondary structure of residues influenced the accuracy of the residue position ( Cα RMSD ) in the models ., In the figure , for example , “HC” indicates cases where the native residue is helix and the modeled residue is coil ., It turned out that correctly predicted helix residues ( class “HH” ) have lower mean Cα RMSD , e . g . are more accurate , than other classes ( one-way ANOVA p = 1 × 10−35 and Tukey’s range test ) ., Next , in S6 Fig , we addressed the influence of the secondary structure prediction agreement on the Cα RMSD of residues ., The X-axis shows the number of secondary structure prediction methods that agree ( e . g . consensus ) on the correct secondary structure of residues and the Y-axis is the Cα RMSD of residues in the models ., Residues where none of the four secondary structure prediction methods predict the correct secondary structure ( consensus, 0 ) have higher ( worse ) mean Cα RMSD than other residues ( one-way ANOVA p = 1 × 10−11 and Tukey’s range test ) ., Thus , we see some influence of the accuracy of predicted secondary structure to the quality of the final model with statistical significance , but as seen from the figures , difference was not very large ., In IDP-LZerD , the fragment generation procedure creates fragments of all three secondary structure classes even if none of the methods predict the correct class to minimize the impact of incorrect secondary structure prediction ., To further examine performance of IDP-LZerD , we compared modeling results with other methods ., While no other methods are designed to model complexes involving long IDPs , some peptide-protein modeling software can use relatively long peptides ., We compared IDP-LZerD with CABS-dock 27 and pepATTRACT 26 , because as seen in Table 1 , these two do not require the binding site as input and the programs are available for us to run ., The CABS-dock web server outputs 10 docking models for a peptide up to 30 amino acids while the pepATTRACT web server outputs 50 docking models and does not explicitly limit the length of the peptide ., The performance was compared on the eleven bound and unbound complexes with IDPs up to 30 amino acids in Tables 2 and 3 ., Within the top 10 , CABS-dock had hits for six bound cases and four unbound cases , pepATTRACT had hits for three bound cases and one unbound case , and IDP-LZerD had hits for seven bound and four unbound cases ( Table 9 ) ., The longest IDP successfully modeled by CABS-dock was 26 amino acids and the longest IDP successfully modeled by pepATTRACT was 22 amino acids ., In contrast , IDP-LZerD had top 10 hits for the longest IDPs in this table ( 27 amino acids; Table, 9 ) in addition to even longer IDPs in the full dataset ( Tables 7 and 8 ) ., Therefore , overall IDP-LZerD showed better performance than the two methods compared ., In addition , we compared the performance of IDP-LZerD to the previously published results of MD-based peptide-protein modeling methods 17–19 ., The protein-peptide complexes used in their literature range from 2–15 amino acids ., Among their datasets , we ran IDP-LZerD on all cases with 11 or more amino acids and unbound receptors , for a total of eight cases ( S6 Table ) ., IDP-LZerD produced acceptable models in the top 10 for five out of eight cases with a sixth case having an acceptable model at rank 306 ( Table 10 ) ., For the two cases with no hits , 2am9 and 1b9k , paths with 5 Å RMSD were created in Step 3 ( Fig, 1 ) but not selected for refinement ., IDP-LZerD and AnchorDock produced the same number of hits , but the models produced by AnchorDock have a lower RMSD ., The results indicate an advantage of MD over coarse-grained approach for short peptides ., They also suggest a potential improvement of IDP-LZerD by employing MD for the initial fragment-docking step , although it would take significantly more computational time than the current procedure ., In addition to the other successful cases , we chose four cases to discuss , which illustrate the usefulness of IDP-LZerD models ., In some disordered PPIs , the IDP forms secondary structure in the bound form that is not seen in isolation ., The interaction between β-catenin and Transcription factor 7-like 2 ( TCF7L2 ) , which is involved in the Wnt signal transduction pathway , is such an example ., In isolation , TCF7L2 exhibits circular dichroism spectra consistent with 96% random coil and 4% β-sheet , indicating that it is intrinsically disordered 57 ., In contrast , the crystal structure of the complex ( 1jpw ) shows a C-terminal helix ( residues 40–50 ) , which was correctly predicted by the secondary structure methods and many models by IDP-LZerD ., For both bound ( 1jpw ) and unbound ( 2z6h ) receptors , the overall complex was well-modeled ( RMSD at the interface , I-RMSD: 2 . 85 Å for bound and 4 . 50 Å for unbound ) with the structure and location of the C-terminal helix and hotspot residue Leu48 ( full atom L-RMSD 1 . 43 Å ) predicted very well in the bound case ( Fig 7A ) ., Interestingly , among 1000 docking models generated , Leu48 was the most frequent contact in both the bound and unbound cases , appearing in 956 models for bound and 944 models for unbound , compared to an average of 685 and 696 , respectively ( S7 Fig ) ., There are two more experimentally verified hotspot residues in the IDP , Glu17 and Asp16 57 ., Glu17 was in contact with the receptor in both bound and unbound cases in more than the average number of models , but Asp16 did not stand out ( S7 Fig ) ., The next examples are complexes between CREB-binding protein ( Cbp ) /p300 TAZ1 domain and its disordered regulator proteins , hypoxia inducible factor 1-α ( HIF-1α ) and its competitive inhibitor , Cbp/p300-interacting transactivator 2 ( CITED2 ) ., HIF-1α and CITED2 are different lengths , have only 12 . 5% sequence identity , and bind differently to the Cbp/p300 TAZ1 domain ( in Fig 7B , the N-terminus of CITED2 is at the bottom right while in panel D the N-terminus of HIF-1α is at the bottom left . The TAZ1 domain is shown in the same orientation in all panels ) ., Nevertheless , the IDPs share a conserved binding motif ( LPEL in CITED2 , LPQL in Hif-1α , referred as LPXL ) 58 ., We docked two complexes: CITED2 with human TAZ1 ( bound , 1p4q ) and HIF-1α with mouse TAZ1 ( bound , 1l8c; unbound , 1u2nA ) ., Because the human TAZ1 domain does not have an available unbound structure , we used its structure in complex with HIF-1α ( 1l3eB ) for the unbound case , which has a binding site RMSD of 5 . 11 Å to the bound form with CITED2 ., Remarkably , the prediction was accurate not only for the bound ( Fig 7B ) , but also for the unbound case ( Fig 7C ) ., Both leucines in the LPXL motif , Leu243 and Leu246 , were experimentally verified as hotspot residues by mutagenesis 59 , but differ in contact consensus among the 1000 models ., Leu243 has above-average counts ( rank 11 , 814 models , average 679 for bound and rank 8 , 851 models , average 713 for unbound ) while Leu246 has below-average counts ( rank 36 , 571 models for bound and rank 40 , 486 models for unbound; S8 Fig ) ., For the mouse homolog , the bound case had no model under 12 . 6 Å L-RMSD in the top 10 ., The rank 16 model shown had L-RMSD 20 . 1 Å , but the LPXL motif is located roughly at the correct position ( Fig 7D ) ., The unbound case had no model with L-RMSD under 10 . 4 Å in the top 10 ., However , HIF-1α was bound to almost the right location in the rank 9 model ( Fig 7E ) , where the fraction of correctly placed ligand residues was 0 . 71 and the L-RMSD of the LPXL motif was 3 . 7 Å ., In addition , the residue Leu795 , which was experimentally determined to be a hotspot residue 60 , has high contact consensus for both bound and unbound ( rank 5 , 911 models , average 734 for bound and rank 8 , 881 models , average 694 for unbound; S9 Fig ) in the final 1000 models ., Thus , in these four models the IDPs were bound almost at the correct place with the LXPL motif predicted particularly well ., Finally , we discuss two cases where predictions did not yield acceptable quality models ., The first case is the complex between Bcl2-like protein 1 ( Bcl2-L-1 ) and Bcl2-associated Antagonist of cell Death ( BAD ) ., While the bound receptor had an excellent result with a medium quality model at rank 1 ( 2bzw; Table 7 , Fig 4a ) , the unbound receptor ( 1pq0 ) had no hits ., However , visual inspection of the top-ranked models shows that the rank 1 to 7 models have a correct IDP conformation and binding site; however , the IDP is rotated by 180° within the binding site ( Fig 4b ) ., Thus , the scoring functions detected a region of affinity but lacked the specificity to distinguish the correct orientation ., The last example is a complex between botulinum neurotoxin type A ( BoNT/A ) and the N-terminal SNARE domain of SNAP25 ( sn2 ) ., BoNT/A causes paralysis by cleaving SNARE proteins which impairs neuronal exocytosis 61 ., Using the bound receptor ( PDB ID: 1xtgA ) , the structure was correctly predicted at rank 5 ( Fig 6d ) ., However , with the unbound receptor ( 1xtfA ) , no hits were found ., In the rank 1 model of the unbound case , while the IDP shows a substantial registration shift , the model occupies 32 . 6% of the binding groove ( top in Fig 7F; measured by the number of receptor residues within 5 Å of both IDPs ) ., Thus , even in cases where no hits are produced , the produced models are reasonable and capture characteristic binding modes of IDPs on their receptors ., The current study presents for the first time that PPIs with long IDPs can be modeled with reasonable accuracy ., By taking advantage of the crucial observation that disordered proteins tend to bind in continuous segments , the procedure is not only more computationally feasible but also functions similarly to the biophysical mechanism of IDP association ., The prediction by IDP-LZerD was successful for the majority of the complexes tested , including unbound cases ., The study further observed that the correct conformation of IDPs are often identified by evaluating docking scores with receptor proteins ., A major challenge in modeling IDP interactions is the existence of fuzziness , where the IDP continues to exhibit multiple conformations in the bound state 11 ., Two cases in the dataset we used are listed as fuzzy complexes in the FuzDB 11: 1g0v ( FuzDB ID FC0018 ) and 3wn7 ( FuzDB ID FC0076 ) ., IDP-LZerD managed to obtain a rank 1 medium hit for 1g0v ( Table 8 ) , while for 3wn7 IDP-LZerD produced an acceptable model at a low rank ., It is particularly challenging to predict complexes where an IDP binds with two or more regions separated by loop regions that do not have direct contact its receptor ( clamp complexes 10 ) , because IDP-LZerD is based on the assumption that each segment of the IDP is in contact with the receptor ., There are several other potential areas of improvement for the method ., Docking larger fragments in cases where the structure of the fragments can be predicted with confidence could improve accuracy ., It is also interesting to employ a coarse-grained model such as CABS 62 for generating fragment conformations and for more efficient structure refinement ., In addition , explicit consideration of receptor flexibility could improve performance , although the soft surface representation used by LZerD already accounts for some degree of receptor flexibility ., A key feature would be the ability to handle phosphorylated residues , as IDPs are frequently sites of post-translational modification and some complexes ., This would require consideration of the effect of phosphorylation on secondary structure in addition to modification of the docking and scoring protocols ., Another potential area of improvement is to guide docking by considering known or predicted hotspot residues on both IDPs and receptor proteins ., Methods that could detect hotspots include computational alanine scanning 63 or applying a statistical scoring function 51 , 52 on a per-residue basis ., Alternatively , as we showed in the case studies ( S7 , S8 and S9 Figs ) some promise was shown that hotspot residues could be predicted by taking consensus binding sites from ensembles of docking models ., Accurate detection of hotspot residues could also lead to improved performance for fuzzy complexes , particularly the clamp class where two or more stably bound regions of the IDP are separated by fuzzy regions ., Disordered PPIs are involved in important roles in various pathways and diseases ., Overall , the work opens up a new possibility of modeling disordered protein interactions , providing structural insights for understanding the molecular mechanisms and malfunctions of these interactions , which are difficult to obtain by both experimental means and conventional computational protein docking methods ., Protein complexes containing IDPs with diverse functions and lengths were selected for developing and testing IDP-LZerD ., Candidate complexes were found from reviews of disordered protein complexes 2 , 6 ., In addition , cases were found in databases of eukaryotic linear motifs ( ELMs ) 53 and fuzzy complexes ( FuzDB ) 11 ., For each case , disorder was verified by searching the literature for experimental evidence and DisProt 13 for a corresponding entry ( if available ) ., Each PDB file was

Introduction, Results, Discussion, Methods

Disordered protein-protein interactions ( PPIs ) , those involving a folded protein and an intrinsically disordered protein ( IDP ) , are prevalent in the cell , including important signaling and regulatory pathways ., IDPs do not adopt a single dominant structure in isolation but often become ordered upon binding ., To aid understanding of the molecular mechanisms of disordered PPIs , it is crucial to obtain the tertiary structure of the PPIs ., However , experimental methods have difficulty in solving disordered PPIs and existing protein-protein and protein-peptide docking methods are not able to model them ., Here we present a novel computational method , IDP-LZerD , which models the conformation of a disordered PPI by considering the biophysical binding mechanism of an IDP to a structured protein , whereby a local segment of the IDP initiates the interaction and subsequently the remaining IDP regions explore and coalesce around the initial binding site ., On a dataset of 22 disordered PPIs with IDPs up to 69 amino acids , successful predictions were made for 21 bound and 18 unbound receptors ., The successful modeling provides additional support for biophysical principles ., Moreover , the new technique significantly expands the capability of protein structure modeling and provides crucial insights into the molecular mechanisms of disordered PPIs .

A substantial fraction of the proteins encoded in genomes are intrinsically disordered proteins ( IDPs ) , which lack a single stable structure in the native state ., IDPs serve many functions including mediating protein-protein interactions ( PPIs ) ., Such disordered PPIs are prevalent in important regulatory pathways , including many interactions of the tumor suppressor protein p53 ., To elucidate the molecular mechanisms of disordered PPIs , obtaining tertiary structure information is essential; however , they are difficult to study with experimental techniques and existing computational protein-protein and protein-peptide modeling methods are unable to model disordered PPIs ., Here we present a novel computational method for modeling the structure of disordered PPIs , which is the first of this sort ., The method , IDP-LZerD , is designed to follow a known biophysical picture of the mechanism of how IDPs interact with structured proteins ., IDP-LZerD successfully modeled the majority of disordered PPIs tested ., This technique opens up new possibilities for structural studies of IDPs and their interactions .

molecular dynamics, protein structure prediction, protein structure, intrinsically disordered proteins, research and analysis methods, protein structure determination, proteins, biological databases, structural proteins, chemistry, proteomics, biophysics, molecular biology, physics, biochemistry, proteomic databases, database and informatics methods, biology and life sciences, physical sciences, computational chemistry, macromolecular structure analysis

journal.ppat.1000701

Direct Restriction of Virus Release and Incorporation of the Interferon-Induced Protein BST-2 into HIV-1 Particles

The innate defense against viruses includes specific host cell proteins with intrinsic abilities to restrict viral replication ., In some cases these restriction factors have been linked to classic aspects of the innate immune response , such as the antiviral state induced by type I interferons ., To replicate in this hostile environment , viruses encode specific antagonists of restriction factors 1 ., Several of the so-called accessory proteins of primate immunodeficiency viruses have been recognized as such antagonists ., For example , the HIV-1 accessory protein Vpu was long known to enhance the release of progeny virions from infected cells , potentially by antagonizing an intrinsic cellular restriction to virion-release 2 , 3 ., The study of this phenomenon led to the discovery of the antiviral activity of a protein with no previously known function , BST-2 ( also known as HM1 . 24 and CD317 ) , now referred to as a “tetherin” based on its ability to retain nascent virions on the surface of infected cells 4–6 ., BST-2 is an interferon-induced , transmembrane and GPI-anchored protein that restricts the release of a number of enveloped viruses including all retroviruses tested as well as members of the arenavirus ( Lassa ) and filovirus ( Ebola and Marburg ) families 7–10 ., However , how BST-2 mediates the retention of nascent virions is unclear ., Viral antagonists of BST-2 include the HIV-1 Vpu , HIV-2 Env , SIV Nef , Ebola glycoprotein , and KSHV K5 proteins 4 , 5 , 11–14 ., A common feature of the antagonism of BST-2 by viral gene products is its removal from the cell surface , the presumed site of virion-tethering activity ., An unusual membrane topology , localization in cholesterol enriched membrane microdomains , and homo-dimerization may each be key to BST-2s restrictive activity ., BST-2 binds the lipid bilayer twice , via both an N-terminal transmembrane domain and a C-terminal GPI anchor 8 ., This topology leads to the hypothesis that BST-2 retains virions by directly spanning the lipid bilayers of the virion and host cell ., Many enveloped viruses including HIV-1 and Ebola bud from cholesterol-enriched membrane domains in which BST-2 is enriched 15 , 16 ., These observations lead to the hypothesis that BST-2 is incorporated into virions as part of the mechanism of restriction ., BST-2 forms disulfide-linked dimers 6 ., This observation leads to the hypothesis that the molecular topology of restriction includes dimerization between virion- and cell-associated BST-2 ., Here , we show that BST-2 is positioned to directly retain virions on the surface of infected cells and is incorporated into virions ., We confirm that virions retained on the cell surface can be released by proteolysis , but find that they are not released by cleavage of GPI-anchors with phosphatidyl inositol specific phospholipase C or by disulfide reduction with dithiothreitol ., Although these findings leave the precise configuration of the BST-2 molecules that restrict release unsolved , they support a model in which BST-2 incorporates into virions to directly restrict their release from the plasma membrane ., This mechanism may be broadly applicable to the inhibition of enveloped viruses by BST-2 ., To test the hypothesis that BST-2 is positioned along the plasma membrane appropriately to directly tether virions , we visualized the location of the molecule using correlative fluorescence and electron microscopy ., To accomplish this , we labeled the surface of HeLa cells , which express BST-2 constitutively 5 , with a specific antibody that recognizes the BST-2 ectodomain 17 ., This antibody was secondarily labeled with cadmium selenide/zinc sulfide nanocrystals ( QDots ) that are both fluorescent and electron dense; this property allowed cells labeled identically to be observed by either light or electron microscopy 18 ., The surfaces of cells labeled for BST-2 were characterized by a punctate staining pattern when visualized by fluorescence microscopy ( Figure 1A and Figures S1 and S2 ) ., This pattern has been noted previously using routine fluorophores 9 , 19 ., In cells expressing HIV-1 , these puncta appear to contain Gag as well as BST-2 and have been hypothesized to reflect sites of virion-formation; in uninfected cells their identity is unclear ., Here , in cultures transfected to express the complete HIV-1 genome including the BST-2 antagonist gene vpu , some cells were characterized by reduced or absent surface staining ( Figures 1B and S1 ) ., In particular , multinucleated cells resulting from virally induced cell-cell fusion were strikingly low in surface BST-2 ( Figures 1B and S1 ) , consistent with the previously described reduction in the expression of cell-surface BST-2 induced by Vpu 5 ., In contrast , no reduction in surface stain was seen when cells were transfected to express a vpu-negative HIV-1 genome; in this case , multinucleated cells resulting from virally induced cell-cell fusion expressed abundant BST-2 on their surfaces ( Figures 1C and S1 ) ., Together , these data indicated that the QDot-based stain for BST-2 revealed the previously noted punctate surface pattern , and it faithfully revealed the removal of BST-2 from the cell surface by Vpu as expressed in the context of the complete viral genome ., To determine the distribution of BST-2 at the ultrastructural level and in relation to nascent HIV virions , cells stained in an identical manner to those shown in Figure 1A-C were processed for transmission electron microscopy ., In uninfected cells , BST-2 was found in foci along the plasma membrane ( Figures 1D and S2 ) , which likely correspond to the puncta seen using immunofluorescence ., Some of these foci were associated with endocytic pits , which appeared either coated or uncoated , whereas other foci were not associated with any apparent structure ., In cells expressing the complete HIV-1 genome including vpu , surface labeling was often relatively sparse , even in areas of clustered viral particles ( Figure 2A ) ., Such paucity of label is consistent with the reduced surface expression visualized by fluorescence microscopy ., These wild type viral particles showed both immature ( crescentic electron density along the perimeter of the budding virion ) , as well as mature morphology ( electron dense cores with occasional conical shape ) ., In contrast , in cells expressing vpu-negative virus , BST-2 was readily detectable at the cell surface ( Figure 2B ) ., Furthermore , label was intercalated between the plasma membrane and nascent virions as well as between nascent virions found in clusters , most of which had a mature morphology ., Occasionally , striking examples of label concentrated at the neck of budding virions in the case of vpu-negative virus were observed ( Figure 2B , inset ) ., These electron microscopic data indicated that BST-2 is positioned appropriately to function as a direct tethering factor ., To determine whether BST-2 is incorporated into virions , we looked for profiles of budding virions and for profiles of virions distant from the cell surface ., Surprisingly , wild type virions were not infrequently labeled for BST-2 ( Figure 2C and E; see Figure S3 for control stain ) ., This result is consistent with functional data indicating that Vpu is not a fully effective antagonist of BST-2 5 , and it is consistent with the virion-capture and immunoblot experiments described below ., In rare cases , label for BST-2 was found directly between virions that appeared linked to each other ( Figure 2E ) ., In the case of vpu-negative virus , label was particularly evident among and between virions caught at a distance from the plasma membrane ( Figure 2D and F ) ., Potential association of such label with the plasma membrane was excluded by electron tomography of thick sections; reconstructed three-dimensional images confirmed the presence of labeled virions that were unambiguously discontinuous with the plasma membrane ( Figure 2G and Video S1 ) ., Although substantial variability was observed in the density of label for BST-2 on and between individual virions , visual inspection of 38 images yielded 358 virions with 149 virion-associated Qdots in the case of wild type virus and 327 virions with 302 associated Qdots in the case of vpu-negative virus , indicating a 2 . 3-fold greater association of label with virions in the absence of vpu ., These immuno-electron microscopic data indicated that BST-2 is incorporated into virions ., The data were also consistent with a model of viral antagonism in which Vpu decreases the density of BST-2 at sites of virion assembly and within virions themselves ., To validate the incorporation of BST-2 into virions , we devised a bead-based virion-capture assay using the same antibody as was used above for the morphologic studies ., A key feature of this assay is the virologic readout of infectivity , allowing confirmation that BST-2 is incorporated into bona fide infectious virions ( Figure 3A-C ) ., Preparations of cell-free virions produced from BST-2-positive HeLa cells were mixed with antibody to the BST-2 ectodomain , or with an isotype-matched control antibody , an antibody to CD44 , or an antibody to CD45 ., The virion-antibody complexes were then captured on coated magnetic microbeads and used to infect adherent CD4-positive HeLa indicator cells in an infectious center assay of HIV-1 infectivity ( Figure 3A-C ) ., CD44 is incorporated into virions and served as a positive control for the capture 20 ., CD45 is excluded from virions produced from hematopoietic cells 21 , but here it serves only as a second specificity control , since CD45 is not known to be expressed on HeLa cells ., Strikingly , antibody to BST-2 captured infectious virus from solution , both in the case of wild type and vpu-negative genomes ., In contrast , infectious virus ( wild type ) produced from HEK293T cells , which do not express BST-2 constitutively 4 , 5 , was not captured by antibody to BST-2 ( Figure 3D ) ., Capture of virions produced from HeLa cells by antibody to BST-2 was confirmed by measurement of viral capsid ( p24 ) antigen by ELISA ( Figure 3E ) ., The efficiency of capture as measured by infectivity or p24 ELISA was not significantly affected by Vpu; this suggests either that Vpu does not significantly decrease the incorporation of BST-2 into virions or that both wild type and vpu-negative virions incorporate a threshold amount of BST-2 sufficient for capture ., Immuno-capture of three independent sets of wild type and vpu-negative virus preparations confirmed the incorporation of BST-2 into infectious virions of HIV-1 ( Figure 3 and data not shown ) ., Immunoblot analysis also supported the conclusion that BST-2 is incorporated into virions and further suggested that Vpu inhibits this ( Figure 4 , in which virions produced from HeLa cells and concentrated by centrifugation were analyzed ) ., Remarkably , when normalized by the volume of the original culture supernates ( Figure 4A , left panel ) , preparations of wild type virions contained more BST-2 , as well as more p24 capsid protein , than virions produced by vpu-negative virus ., This difference in BST-2 contents in the volume-normalized samples suggests that the signal was derived from virions and not merely cellular debris or exosomes; if the latter were the case , then the volume normalized samples from cultures expressing wild type virus should have contained less BST-2 , due to Vpu-mediated down-regulation ., In contrast , when the preparations of virions were normalized by their contents of p24 antigen , BST-2 was essentially only detectable in the absence of Vpu ( Figure 4A and B ) ., The apparent association of BST-2 with virions and a relative decrease in the content of BST-2 in virions produced in the presence of Vpu was observed in three independent preparations ., These observations were robust to filtration of the virion preparations through 0 . 22 µM pore size membranes , suggesting that the detection of BST-2 was not due to the presence of aggregates of BST-2-containing cellular vesicles and virions ( data not shown ) ., Interestingly , the relatively greater phenotype of vpu detected in this assay ( an apparent 40-fold increase in virion output ) as compared to that detected by measurement of p24 in unfractionated culture supernatants by ELISA ( a 5–8-fold increase; see Figures 5 and 6 ) may be due to a reduced fraction of pelletable p24 when virions are produced in the absence of Vpu ( data not shown ) ., Notably , virions produced in the absence of Vpu contained , in addition to a triplet of species that migrated with apparent molecular mass in the range of 27–37 kDa ( likely representing heterogeneously glycosylated BST-2 ) , two bands of under 20 kDa in apparent mass ., These species are less than the predicted size of unmodified BST-2 ( 20 kDa ) , and their identity is unknown; conceivably , they could represent proteolysis of higher mass forms ., Overall , these immunoblot data , like the results of immuno-electron microscopy and immuno-capture , support the conclusion that BST-2 is incorporated into virions ., Furthermore , the immunoblot results suggest that Vpu reduces the virion-incorporation of BST-2 ., To support further a direct tethering model , we confirmed that proteolysis with subtilisin releases virions retained on the cell surface Figure 5A , in which the black bars indicate the fraction of the total amount of p24 capsid antigen produced by the culture that was spontaneously released into the medium after transfection with wild type or vpu-negative ( Δvpu ) proviral plasmids; the dark gray bars indicate the fraction of the total that was further eluted from the cells with buffer ( control ) or subtilisin; and the light gray bars indicate the fraction of the cell-associated p24 that was eluted with buffer or subtilisin 22 ., The fractional elution of p24 was greater in the absence of Vpu , consistent with a greater number of virions initially retained at the cell surface ., Notably , these quantitative data indicated that the total fraction of p24 releasable from the cells ( adding that released spontaneously to that released by proteolysis with subtilisin; open bars in Figure 5A ) is greater in the case of wild type than Δvpu ., The “non-releasable” p24 in the case of cells expressing vpu-negative virus presumably reflects virions that have been endocytosed subsequent to restricted release and are not accessible to proteolysis ., We further showed that proteolysis with subtilisin indeed acts on BST-2; it largely removed the BST-2 ectodomain from the cell surface as detected by flow cytometry ( Figure 5B ) , and it degraded the ectodomain in vitro ( Figure 5C ) ., These results are consistent with direct tethering mediated by BST-2 , but they do not discriminate among several potential topological models of restriction ( Figure 5 , D-F ) ., The preceding data suggest that the incorporation of BST-2 into viral envelopes and a direct tethering mechanism underlie its restrictive activity ., However , the topology of the BST-2 molecules that mediate the retention of nascent virions remained unclear ., One hypothesis is that virion-associated BST-2 interacts directly with cell surface BST-2 , potentially via disulfide bonds but alternatively via predicted coiled-coil regions in the ectodomain of the protein ( Figure 5D and 4 ) ., Alternatively , one end of BST-2 could embed in the lipid bilayer of the cell and the other in that of the virion ., Such membrane-spanning models are depicted in Figure 5E and F; notably , BST-2 dimers could span the virion and host membranes in parallel or anti-parallel orientations ., Here , release of nascent virions was not obtained by incubation of virus-producing cells with phosphatidyl inositol ( PI ) -specific phospholipase C ( PLC ) , which is expected to cleave the GPI-anchor of BST-2 ( Figure 6A ) ., Because BST-2 remains attached to the cell surface by its transmembrane domain after cleavage of its GPI anchor ( data not shown ) , PI-PLC activity was validated using CD55 ( decay accelerating factor ) , which is a typical GPI-anchored protein ( Figure 6D; in which PI-PLC effectively removed CD55 from the cell surface ) ., These data weighed against the membrane-spanning parallel dimer model of Figure 5F ., Incubation of cells with dithiothreitol ( DTT ) to reduce disulfide bonds also failed to release virions ( Figure 6B ) , weighing against a self-interaction mechanism mediated exclusively by disulfide bonds ., Incubation with PI-PLC followed by DTT ( Figure 6C ) also failed to release virions , weighing against an anti-parallel , disulfide linked , membrane-spanning model ( Figure 5E ) ., These data do not provide direct support for any specific topology of restriction , but they leave open the possibility that the model shown in Figure 5D is operative via a coiled-coil based interaction between the ectodomains of virion- and cell-associated BST-2 ., The interferon induced , GPI-anchored and transmembrane protein BST-2 restricts the release of enveloped virions from infected cells by an unclear mechanism ., Here , the prototypic restricted virus , HIV-1 , and the prototypic viral antagonist protein , Vpu , were used to investigate this mechanism ., The data provide key initial support for a model in which BST-2 is a direct tethering factor that is itself incorporated into infectious virions ., Recent reports have questioned the incorporation of BST-2 into virions and the co-localization of BST-2 with virion proteins , leaving a direct tethering model of restriction unsupported 23 , 24 ., An inability to detect BST-2 in virions by immunoblot may be attributable to insufficient sensitivity of the assay , whereas it is more difficult to explain the reported negative data for co-localization ., Here , a combination of morphologic , virologic , and biochemical approaches provided evidence supporting direct tethering and virion-incorporation of BST-2 ., Evidence that BST-2 is incorporated into virions was provided by immuno-electron microscopy , immuno-capture of infectious virions , and routine immunoblot ., The immuno-electron microscopic data specifically localized BST-2 as adjacent to virions , between virions and the plasma membrane , and in rare instances between virions that were linked to each other ., The electron microscopic data also suggested that the punctate distribution of BST-2 seen at the cell surface by fluorescence microscopy is only partly due to the occurrence of the protein in endocytic pits ., Many of the foci seen along the plasma membrane were not associated with any apparent structure ., Intriguingly , these foci could represent membrane microdomains containing BST-2 , such as cholesterol-enriched lipid rafts , although we cannot exclude that they reflect antibody-induced lateral aggregation of BST-2 within the lipid bilayer ., Somewhat surprisingly , immuno-electron microscopy , immuno-capture of infectious virions , and routine immunoblot each indicated that virions produced in the presence of Vpu are not devoid of BST-2 ., However , immunoblot , and to a lesser extent electron microscopy , suggested that Vpu decreases the amount of BST-2 in virions ., Notably , antibody to the BST-2 ectodomain captured virions produced in the presence or absence of Vpu equally well; this may reflect a threshold amount of virion-associated BST-2 required for immuno-capture that is met by virions produced in either context ., Altogether , these data indicate the presence of BST-2 in virions ., The data also support a relative but not absolute exclusion of BST-2 from virions by Vpu ., One of two topological models has seemed likely to explain restriction mediated directly by BST-2: a membrane spanning model in which BST-2 embeds one end in the cell membrane and the other in the virion membrane , or a self-interaction model in which virion-associated and cell-surface-associated BST-2 molecules interact via their ectodomains ., Here , we found no direct support for membrane spanning models; phosphatidyl inositol-specific phospholipase C ( PI-PLC ) , either with or without disulfide reduction , failed to release virions retained on the surface of BST-2-expressing cells ., These results weigh against membrane spanning models involving parallel dimers or anti-parallel dimers held together by disulfide bonds ., A caveat to this interpretation is that the failure of PI-PLC to release virions could be due to modification of the GPI anchor of BST-2 such that it is resistant to this enzyme , a possibility not excluded by the enzyme activity control herein ( release of CD55 ) ., As noted above , PI-PLC did not release BST-2 from the cell surface , consistent with membrane anchoring by the proteins transmembrane domain ( data not shown ) ; consequently , we could not validate the activity of PI-PLC on native BST-2 ., Similarly , the failure of reduction with DTT to release virions could be due to an inability of DTT to reach the key disulfide bond ( s ) at physiological temperature , for example , if they are protected within a well-folded structure ., Similar attempts to release virus-like-particles of Ebola retained on the cell surface by BST-2 with as much as 500 mM DTT were also ineffective 14 ., Recent mutational analyses of cysteine residues within the BST-2 ectodomain suggest that disulfide-mediated dimerization is a correlate of the restriction of HIV-1 release , but not of arenavirus release 25 , 26 ., Notwithstanding these potentially conflicting findings , exactly how disulfide mediated dimerization would contribute to restriction , if not by mediating an interaction between virion- and cell-associated BST-2 , is unclear ., A direct restriction mechanism that is not disfavored by any of the data herein is an ectodomain self-interaction model such as that shown in Figure 5D , but in which the driving force of tethering is a coiled-coil based interaction ., To test this hypothesis , the role of the predicted coiled-coil region in the ectodomain needs to be directly and critically evaluated: are key residues in the predicted structure required for restriction ?, Does the putative interaction responsible for restriction involve two , three , four , or more α-helices ?, Although each of the models of Figure 5 may be too simplistic , a self-interaction model of restriction is attractive: a single plasma membrane protein , BST-2 , is localized to sites of viral assembly , incorporates into virions , and dimerizes or forms higher order multimers or aggregates to restrict release ., This direct tethering , self-interaction model of restriction relies only on the localization of BST-2 to sites of viral budding and on the incorporation of BST-2 into virions ., Consequently , it can potentially be generalized to all enveloped viruses that assemble on membrane microdomains that contain BST-2 ., Conversely , a model of relief of restriction by removal of BST-2 from the sites of viral assembly and from virions themselves can potentially be generalized to all viral proteins that decrease the expression of BST-2 at the plasma membrane ., So far , such proteins include HIV-1 Vpu , HIV-2 Env , SIV Nef , and KSHV K5 5 , 11–13 ., Notably , the data herein indicate the presence of BST-2 in infectious HIV-1 virions that are spontaneously released from cells , even when the viral antagonist protein Vpu is expressed ., The HeLa cells used for the studies herein express BST-2 endogenously , obviating transient expression methods that may be prone to artifactual over-expression of BST-2 in individual cells ., On the other hand , whether wild-type virions produced from primary T cells or produced in vivo contain BST-2 remains to be determined ., The observations herein suggest the possibility that virion-associated BST-2 serves functions in addition to tethering nascent virions ., In this regard , we note the potential for virion-associated BST-2 to interact with ligands , including itself , on immune effector cells ., BST-2 is constitutively expressed , at least in mice , on plasmacytoid dendritic cells ( PDCs ) , as well as on B cells and activated T cells in humans 7 , 27 ., Considering the incorporation of BST-2 into virions and the potential for interaction between virion- and cell-associated BST-2 , we speculate that in addition to its direct antiviral activity as a tetherin , BST-2 might flag enveloped viruses for subsequent binding to PDCs and B-cells , which are antigen-presenting cells , and so stimulate the host adaptive immune response ., Recently , BST-2 was identified as a ligand for a receptor on PDCs , ILT7 , which transduces a signal for shut-off of interferon production 28 ., Based in these findings and the data herein , we also speculate that virion-associated BST-2 might provide negative feedback to the interferon response ., These mechanisms would place BST-2 at the interface of innate and adaptive immunity to enveloped viruses ., In conclusion , the data herein advance a direct model of restriction in which BST-2 is found both at the sites of viral assembly along the plasma membrane and within budding and nascent virions ., While this paper was being finalized , independent evidence for direct restriction of virus release and virion-incorporation of BST-2 was reported 29 ., Biochemical data suggested a parallel dimer configuration ., Strikingly , an “artificial tetherin” that lacks primary sequence homology with BST-2 , but which contains its key membrane binding and structural features , showed antiviral activity , indicating that no cellular cofactors are likely obligatory for the tethering phenomenon 29 ., Directly relevant to the data herein , a mutant BST-2 lacking a GPI-anchor site was incorporated into virions but was unable to restrict virion release ., These observations would leave open the role of the GPI-anchor in restriction , if not as one of the two membrane anchors in virion-cell membrane spanning models ., On the other hand , this study directly demonstrated a requirement for the coiled-coil ectodomain of BST-2 in restriction and showed that a heterologous , dimeric coiled-coil could provide restrictive activity ., These data can be interpreted to support a coiled-coil based self-interaction model of restriction ., Alternatively , as proposed by the authors , the coiled-coil structure could be needed for an extended conformation of the ectodomain , which might facilitate spanning of the virion- and cell-membranes 29 ., While the molecular topology of the BST-2 molecules that restrict virion release thus remains to be resolved , the augmentation of BST-2 activity and the inhibition of viral antagonists such as Vpu likely represent new approaches to the prevention and treatment of infections due to enveloped viruses ., The development of these approaches depends on understanding the regulation of BST-2 during the immune response as well as on deciphering the structural basis of virion tethering and of the action of viral proteins that antagonize BST-2 function ., The proviral plasmid pNL4–3 was obtained from the National Institutes of Health ( NIH ) AIDS Research & Reference Reagent Program and contributed by Malcolm Martin 30 ., The pNL4-3 mutant ΔVpu ( vpuDEL-1 ) was provided by Klaus Strebel 31 ., The murine monoclonal antibody to BST-2/HM1 . 24/CD317 and the BST-2 ectodomain protein were gifts from Chugai Pharmaceutical Co . , Kanagawa , Japan 17 ., For flow cytometry , an IgG2a antibody isotype control , a goat , anti-mouse IgG antibody conjugated to allophycocyanin ( APC ) and a FITC-conjugated antibody to CD55/DAF were obtained from BioLegend ( San Diego , CA ) ., For immunofluorescence and immuno-electron microscopy , a goat anti-mouse IgG antibody conjugated to biotin was obtained from Jackson ImmunoResearch ( West Grove , PA ) , and streptavidin-conjugated cadmium selenide/zinc sulfide nanocrystals ( quantum dots; QDot 625 ) were obtained from Invitrogen ( Carlsbad , CA ) ., Subtilisin was from Sigma-Aldrich ., PI-specific phospholipase-C was from Prozyme , San Leandro , CA or Sigma-Aldrich ., The HeLa cells used in this study were clone P4 . R5 , which express both CD4 and CCR5 and were obtained from Ned Landau; these cells are a derivative of clone P4 and were maintained in DMEM plus 10% fetal bovine serum ( FBS ) , penicillin/streptomycin , and puromycin 32 ., HEK293T cells were also obtained from Ned Landau and were maintained in EMEM plus 10% FBS and L-glutamine ., Cells were transfected using Lipofectamine2000 ( Invitrogen ) according to the manufacturers instructions ., For the microscopic experiments , cells were transfected in MatTek glass bottom dishes , using 0 . 8 µg pNL4-3 or pvpuDel-1 ., For production of virions , cells were plated in 10 cm tissue culture dishes and transfected with 16 µg of pNL4-3 or pvpuDel-1 ., HeLa P4 . R5 cells were plated on coated MatTek glass bottom dishes and transfected as indicated above ., One day after transfection , the cells were fixed using 3% formaldehyde in PBS and stained using the murine monoclonal antibody to BST-2/HM1 . 24/CD317 ( 0 . 05 µg/ml ) , followed by goat anti-mouse-biotin ( 1∶100 ) and streptavidin-QDot 625 ( 1∶100 ) ., For fluorescence microscopy , cells were mounted in anti-fade media containing DAPI , and image data were obtained using an Olympus laser scanning confocal microscope ., Z-series of two-channel images were colored , merged and projected using Image J . For transmission electron microscopy , parallel samples were re-fixed in 2% glutaraldehyde ( EM Sciences ) in 100 mM sodium cacodylate buffer ( pH 7 . 4 ) for 30 min , post-fixed in 1% osmium tetroxide for 30 min , stained in 2% uranyl acetate in water for 1 h , dehydrated in an ethanol gradient , and embedded in Durcupan ACM ( Fluka ) ., Thin sections were stained with Satos lead ., Micrographs were obtained using a JEOL 1200 transmission electron microscope operated at 80kV ., For electron tomography , sections of approximately 250 nm thickness were stained with Satos lead and 2% uranyl acetate ., Series of micrographs were collected on a FEI Titan transmission electron microscope at 300 keV while the sample was tilted from −60° to +60° in 2° increments ., The micrographs were digitized and aligned using IMOD software 33 ., A transform-based back projection software package was then used to create the final alignment and back projection resulting in a three-dimensional volume 34 ., Virus was produced from either HeLa P4 . R5 cells or HEK 293 T cells ., Cells ( 6×106 ) were plated in 10 cm tissue culture dishes and transfected as described above with 16 µg of either pNL4-3 or pvpuDel-1 ., Virus-containing culture supernatants were harvested 48 hours later and clarified by centrifugation at 400×g to remove cellular debris ., Clarified culture supernatants were incubated with antibodies as indicated and then complexed to protein G-coated magnetic microbeads ( Miltenyi Biotec , Bergisch Gladbach , Germany ) according to the manufacturers instructions ., Bead-bound virions were captured using Miltenyi magnetic columns , washed , and eluted using DMEM plus 10%FBS and penicillin/streptomycin/puromycin in the same volume as the input supernatant ., Viral protein levels in the eluate were determined using p24 capsid ELISA ( Perkin-Elmer ) ., Infectious center assays of viral infectivity were performed using HeLaP4 . R5 indicator cells as targets ., Infected foci were developed with X-gal , imaged using a CCD camera , and quantified using image analysis software , as described previously 35 ., For analysis of surface levels of BST-2 , cells were stained before fixation in phosphate buffered saline ( PBS ) including sodium azide and 2% FBS at 4°C using an indirect method to detect BST-2: the HM1 . 24 murine monoclonal antibody ( 0 . 1 µg/ml ) was followed by a goat anti-mouse IgG conjugated to APC ., The gate for BST-2 was set using an antibody isotype control ( IgG2a ) as the primary antibody ., For measurement of CD55 , cells were stained before fixation either with a FITC-conjugated isotype control or a FITC-conjugated anti-CD55 ., Cells were ga

Introduction, Results, Discussion, Materials and Methods

Investigation of the Vpu protein of HIV-1 recently uncovered a novel aspect of the mammalian innate response to enveloped viruses: retention of progeny virions on the surface of infected cells by the interferon-induced , transmembrane and GPI-anchored protein BST-2 ( CD317; tetherin ) ., BST-2 inhibits diverse families of enveloped viruses , but how it restricts viral release is unclear ., Here , immuno-electron microscopic data indicate that BST-2 is positioned to directly retain nascent HIV virions on the plasma membrane of infected cells and is incorporated into virions ., Virion-incorporation was confirmed by capture of infectivity using antibody to the ectodomain of BST-2 ., Consistent with a direct tethering mechanism , we confirmed that proteolysis releases restricted virions and further show that this removed the ectodomain of BST-2 from the cell surface ., Unexpectedly , enzymatic cleavage of GPI anchors did not release restricted virions , weighing against models in which individual BST-2 molecules span the virion and host cell membranes ., Although the exact molecular topology of restriction remains unsolved , we suggest that the incorporation of BST-2 into viral envelopes underlies its broad restrictive activity , whereas its relative exclusion from virions and sites of viral assembly by proteins such as HIV-1 Vpu may provide viral antagonism of restriction .

The cellular protein BST-2 prevents newly formed particles of HIV-1 and other enveloped viruses from escaping the infected cell by an unclear mechanism ., Here , we show that BST-2 is appropriately positioned to directly retain newly formed HIV-1 virions on the cell surface and is incorporated into infectious virions ., We suggest that the incorporation of BST-2 into virions is a key aspect of the proteins broadly restrictive activity against enveloped viruses .

virology/host antiviral responses, biochemistry, microbiology, immunology/innate immunity

journal.pcbi.1005406

A magnesium-induced triplex pre-organizes the SAM-II riboswitch

Non-coding RNAs are currently thought to account for over 75% of the human genome 1 ., In bacteria , non-coding RNAs play important roles in gene regulation ., One such class of RNAs , riboswitches , regulates metabolite production ., Here , a single RNA sequence folds into one of two or more mutually exclusive folds depending on the metabolite concentration 2 , 3 ., In some cases , such as the S-adenosylmethionine-I ( SAM-I ) riboswitch , the RNA contains a transcriptional terminator that forms when ligand is present , in effect silencing genes important for ligand production 4–6 ., When the ligand is not present , the terminator does not form , allowing gene expression , and therefore ligand production , to continue efficiently ., In other cases , such as the SAM-II riboswitch , ligand binding may lead to sequestration of the Shine/Dalgarno sequence , likely blocking ribosome binding and , as a consequence , protein synthesis 7 ., While these examples of ligand-dependent secondary structure switches have been known for some time , a detailed thermodynamic understanding at the atomistic level , including the indispensable effect of the RNA’s ion-atmosphere , has not been achieved ., In recent years , riboswitches have become canonical systems for studies of diverse RNA behaviors , as they possess quintessential characteristics of many RNA systems: ligand binding , Magnesium ion ( Mg2+ ) sensitivity , conformational changes , secondary structure remodeling , and regulatory functions ., Chemical footprinting , NMR , small-angle X-ray scattering ( SAXS ) and single molecule FRET techniques are being exploited to elucidate the folding kinetics , thermodynamics and the magnesium ion sensitivity in RNA systems such as the TPP riboswitch 8 , glycine-dependent riboswitches 9 , different variants of P4-P6 RNA 10–12 and P5abc subdomain of the Tetrahymena group I intron ribozyme 13 ., Other work focuses on more complex functions , such as splicing and ligand recognition and their associations with proteins or different metabolites 14 , 15 ., Success in understanding the structural , dynamical and functional aspects of riboswitch systems requires an integrated experimental and theoretical approach ., Traditional crystallographic techniques produce static snapshots of the riboswitch ., SmFRET , NMR , and SAXS methods obtain kinetic information and overall distributions of conformations ., Molecular simulation allows one to integrate disparate experimental data into a single coherent picture , characterizing transitions in atomistic detail and the free energy landscape with fine resolution ., A large number of riboswitches have been crystallized and have also been investigated via fluorescence and single molecule techniques 16–28 ., Molecular simulations have also been used to study a number of riboswitches , including but not limited to the SAM-I , SAM-II , pre-Q , and adenine riboswitches 29–32 ., Some of these studies replaced the essential atmosphere of divalent Mg2+ ions with other , monovalent ions ., Others used a repulsive Debye-Hückel interaction between the phosphate groups of the RNA ., Such implicit treatments of the ions , however , neglect important near-field effects that occur inside the core of the riboswitch , where the ions are strongly coupled to the RNA ., Although the role of Mg2+ in stabilizing the RNA tertiary structure has long been realized 33 , 34 , the molecular basis of ion-RNA interactions , in terms of structure and function , is not well understood ., In a pioneering study , Draper and co-workers distinguished three classes of ion environments:, ( i ) the diffuse ions , which are not restrained to any particular region ,, ( ii ) water-surrounded ions separated from the RNA by a single hydration layer , which we call outer-sphere ions , and, ( iii ) chelated ions in the inner sphere , which form direct contacts with at least two different phosphate groups of the RNA 34 , 35 ., While the potential role of chelated ions has been emphasized in many studies , recent work using explicit solvent molecular dynamics instead highlights a dense layer of outer-sphere Mg2+ ions , which are primarily responsible for anchoring the RNA structure 36 ., These outer-sphere Mg2+ ions are only transiently bound but nonetheless strongly coupled to the RNA dynamics ., Such highly correlated Mg2+ ions may even reside in the core region of riboswitch RNAs 36 , 37 ., This dynamic cloud of Mg2+ has also been simulated in all-atom reduced models that combine Manning theory with a background of monovalent ions , represented by Debye-Hückel interactions 37 , 38 ., In addition to these native basin simulations , studies of metabolite recognition and specificity have also been initiated using conformational ensemble sampling , again in the absence of Mg2+ ions 39 ., While most riboswitch studies have focused on riboswitches in the 5’-UTR of mRNA that control transcription , less attention has been paid to translational control by riboswitches through ligand-dependent sequestration of the ribosome binding site ( i . e . , the Shine/Dalgarno sequence ) ., The SAM-II riboswitch is one such relatively small RNA element which regulates methionine and SAM biosynthesis ., A single hairpin , classic H-type pseudoknot and triplex interaction near the ligand binding site make this RNA an interesting system to study RNA control over the translation initiation process 21 , 40–44 ., A previous single molecule fluorescence resonance energy transfer ( smFRET ) 21 imaging study sheds light on the dynamic nature of ligand-free SAM-II riboswitch , which becomes conformationally restrained upon ligand binding ., The flexibility of such highly transient conformations is tuned to ensure a viable time scale for conformational transitions in the absence of ligand ., More rapid sensing could , however , be achieved if the riboswitch adopted a binding competent conformation in the ligand-free state ., Mg2+ ions can act as effective anchors , aiding in the preservation of the structural integrity of the RNA ., The emergence of two distinct FRET configurations in the presence of 2 mM Mg2+ in the ligand free system suggests that Mg2+ has the ability to compress the RNA structure , in such a way that it might pre-organize the RNA to form a binding competent conformation ., In a series of small angle X-ray scattering ( SAXS ) experiments , we observed an analogous signature of Mg2+ induced structural collapse that can facilitate subsequent ligand binding 40 ., Our studies provided direct insight into the global rearrangement induced by both Mg2+ and ligand ., The compaction of RNA by Mg2+ was also studied by size-exclusion chromatography ( SEC ) : changes in the measured elution volume suggested a decrease in the particles’ hydrodynamic radius 40 ., Pre-organization by Mg2+ has also been observed in other riboswitches 21 , 28 , 45 ., In the SAM-I system , our biochemical studies have shown that addition of Mg2+ yields the pre-organized partially folded state ., In addition , we have shown that , in the absence of Mg2+ , the fully folded state cannot be achieved , even at high ligand concentrations 23 , 28 , 46 ., The presence of both , Mg2+ and ligand are required for the stabilization of the fully folded ligand-bound configuration ., While previous smFRET and SAXS data revealed that ligand free RNA undergoes substantial structural changes upon variation of Mg2+ concentration , these structural changes often remained undetected by traditional NMR and X-ray crystallography techniques because of the transient nature and low population levels of such intermediates 47 ., The newly developed Chemical Exchange Saturation Transfer ( CEST ) measurements are now capable of probing these sparsely and transiently populated RNA conformations ., Earlier we studied NMR dynamics of the SAM-II system with this new method 47 ., The data indeed confirmed that SAM-II riboswitch can access a sparsely populated but bound-like pre-organized state even in the absence of ligand 40 , 47 ., In the present study , we performed molecular simulations to predict the effect of Mg2+ on the conformational landscape of the SAM-II riboswitch ., We then tested these predictions with 13C-CEST data ., Analysis of our simulations yields the free energy landscape of the SAM-II riboswitch , the effect of Mg2+ on this landscape and insight into the microscopic origins of these effects ., More specifically , reappraisal of the 13C-CEST data for the ligand-free SAM-II riboswitch at different Mg2+ concentrations enabled us to probe the influence of Mg2+ on sparsely populated bound-like pre-organized states ., We then revisited earlier smFRET , SAXS and SEC elution profiles and compared with our present equilibrium simulation results to integrate these data into a unified scenario of Mg2+-induced collapse ., We calculated the free energy landscape of the SAM-II riboswitch using our recently developed all-atom structure-based model ( SBM ) that includes explicit Mg2+ ions and the effects of Manning condensation and Debye-Hückel Potassium and Chloride interactions ., We specifically predict that , as Mg2+ concentration is increased from 0 . 25 mM to 2 mM , the SAM-II riboswitch collapses from an extended , partially unfolded state to a highly compact , pre-organized state , in agreement with the 13C-CEST studies , where we observe a shift in population towards a bound-like conformation ., In addition , our simulations characterize this collapse transition in terms of the radius of gyration as a function of Mg2+ concentration , which is qualitatively similar to previous SAXS measurements ., This agreement gives us confidence in the microscopic details of our simulations , showing that the triplex formation between helix P2b and loop L1 plays an important role in the collapse process ., As mentioned earlier , CEST experiments are able to capture transiently populated dynamic conformations 13 ., This strategy was applied to the ligand-free SAM-II riboswitch in the presence of 0 . 25 mM and 2 mM Mg2+ ., The 13C-CEST profiles of the labeled ribose C1’ and base C6 carbons of C43 were recorded at three different B1-fields ( 17 . 5 , 27 . 9 , 37 . 8 Hz ) with a mixing time of 0 . 3 s at 298 K 47 ., We compare the data for 0 . 25 mM and 2 mM Mg2+ concentrations at B1-field of 17 . 5 Hz ( Fig 2a ) ., The data were fit with a two-state model where the low population of the partially closed state ( peaks around 300 Hz spin-lock offset ) appears to increase with addition of 2 mM Mg2+ ., Consistent results were obtained from CEST profiles for other B1-fields of 27 . 9 and 37 . 8 Hz ( Fig S1 in the S1 Text ) ., Trajectory plots of the fraction of native contacts extracted from the generalized Manning equilibrium simulations of ligand-free SAM-II at these two concentrations clearly show the hopping between different conformations ( Fig 2b ) ., Furthermore , the dynamic transitions between the two major states ( bound-like: Q≈0 . 9 and open: Q≈0 . 7 ) visit native-like conformations ( Fig 2c ) more frequently at 2 mM than at 0 . 25 mM Mg2+ , as summarized in the corresponding contact histograms , P ( Q ) ( Fig 2d ) ., Both CEST experiments and simulation data indicate that the equilibrium shifts from the open conformations toward the native bound-like state as we increase Mg2+ concentration ., The signature of the existence of such Mg2+ induced bound-like states has also been reported in previous smFRET experiments ( Fig 2e ) 22 ., To support our observations we have revisited some of these smFRET efficiency assessments 22 and compared them with theoretical FRET predictions obtained from our generalized Manning model simulations under similar buffer conditions ., The equation used for theoretical FRET prediction is described in section S1 in the S1 Text ., We tracked the dynamics of positions 14 and 52 , where acceptor ( cy5 ) and donor ( cy3 ) fluorophore labels were placed in the smFRET experiments ( Fig 2f ) ., Both experimental and simulation FRET confirm the coexistence of two states at 2 mM Mg2+ ( Fig 2g ) 22 ., Previous SAXS data corroborates well the existing smFRET observations 22 , 40 ., The SAXS data also indicated both ligand and Mg2+ ions are required to effectively fold this riboswitch ., To microscopically understand their mutual and stand-alone effects from the present simulations , we studied the conformational differences of this riboswitch in four extreme buffer conditions and compared our computational results with experimental SAXS data ., For this comparison , we extracted multiple snapshots from several long trajectories and computed ensemble averaged SAXS profiles using the Debye formula for spherical scatterers parameterized in the FoXS web server 48 , 49 as described in section S2 in the S1 Text ., The predicted SAXS curves here show qualitative agreement with experiments ( Fig 2h ) 40 ., The Kratky representation of SAXS data presented in Fig s2 in the S1 Text shows a pronounced peak , indicating the emergence of more extended conformations with decreasing Mg2+ concentrations ., We note that capturing the entire conformational heterogeneity of an extended state is computationally challenging ., This mostly applies for the extreme case where neither ligand nor Mg2+ is present ., In this case , the correlation between theoretical and experimental SAXS profiles leaves room for improvement ., Values for chi-square reflect that and are shown in Table S1 in the S1 Text ., These analyses indeed suggest the potential impact of both , ligand and Mg2+ , stabilizing the closed conformations , which we characterize further below with contact data to describe the pre-organization and the ligand-organized closing ., A significant Mg2+ induced collapse transition , as indicated by the SAXS data , has been followed over a wide concentration range ( up to 100 mM ) of Mg2+ in SEC elution volume profile ( Fig 2i ) ., Here RNA elutes after longer retention times with increasing Mg2+ concentration ( Mg2+ ) in the mobile phase 40 ., Bigger elution volume signifies decreasing hydrodynamic radius of a monomeric RNA molecule ., The folding transitions , both from experimental elution volume data and from average Rg measured from the equilibrium simulation analysis as functions of Mg2+ , follow sigmoid curves with transition midpoint , Mg1/2 at 6 mM ( Fig 2i ) ., At this point , a range of experimental techniques and simulation data support the existence of pre-organized states ., Here we aim to obtain a thermodynamic description of how Mg2+ governs the energy landscape of RNA from our model simulation study ., In Fig 3a , we show the free energy landscape for the folding transition of SAM-II riboswitch in SAM-bound ( in the presence of explicit ligand ) and SAM-free ( in the absence of ligand ) conditions near the physiological concentration of Mg2+ ( Mg2+ = 2 . 0 mM ) ., During this folding transition , each secondary structural segment folds sequentially illuminating the pathway of folding ( Fig 3b ) ., The free energy profile , in the presence of explicit SAM has a distinct bound-state-well , reflecting the ligand-induced stabilization of the closed conformations ( designated as, ( i ) in Fig 3c ) ., In the apo-form of the riboswitch , the fully closed bound state does not correspond to a minimum in the landscape ., At lower Q than this ligand-bound state , the free energy profile for apo-SAM-II riboswitch reveals three distinct minima ., They involve: a ligand-free partially closed state ( state, ( ii ) in Fig 3c ) , which has a substantial overlap with the ligand-bound closed conformation ., In this state , the nonlocal contacts ( involving base-pairing contacts ) including base-stacking contacts in P1 , the P1-L3 pseudo-knot interaction , and major segments of P2b and the L1-P2b triplex interactions remain secured , while the contacts involved in Shine/Dalgarno sequence ( AAAG50G51A523´ ) , and in the part of L1-P2b are disrupted ( state, ( ii ) in Fig 3c ) ., Recent fluorescence and NMR spectroscopic data also indicated that C16 in P2a helix remains mostly unpaired in the absence of SAM 22 ., The data also suggested that formation of the pseudoknot in the absence of SAM is highly transient in nature ., Intermediate states ,, ( iii ) and, ( iv ) in Fig 3c , although marginally separated by a small barrier , effectively belong to a broad , flat basin which involves an ensemble of partially folded open configurations ., A representative unfolded structure (, ( v ) in Fig 3c ) is shown to describe the unfolded minimum ., As we increase the concentration of Mg2+ we find enhanced stabilization of the pre-organized partially closed conformations ( state, ( ii ) in Fig 3d ) relative to the open conformations ., Our latest 13C-CEST chemical exchange data anticipated that the emergence of Mg-induced pre-organization can have immense consequences for rapid ligand recognition 47 ., In the context of the simulation , Mg2+ induced thermodynamic stabilization is reflected by the difference in stability , ΔGPC-PO between the bound-like partially closed ( PC ) conformation and the partially open ( PO ) conformation and by ΔGU-O between the unfolded ( U ) and the open conformation ( O ) ., These two stability differences , ΔGPC-PO and ΔGU-O vary upon increasing Mg2+ until they reach their saturation limits ., Both ΔGPC-PO and ΔGU-O plotted as functions of Mg2+ , are fitted well to sigmoid curves with a Mg2+1/2 value around 6 mM ( inset of Fig 3d ) which again correlates well with the SEC elution volume data ( Fig 2i ) 40 ., To address the open question of how Mg2+ ions regulate structural collapse , we have determined the Mg2+ distribution in the ion-solvation layer of SAM-II , which accommodates increasing numbers of Mg2+ up to 8 mM Mg2+ content ( Fig 4a ) ., Subsequent additions of Mg2+ beyond 8 mM do not effectively add to the 1st layer of Mg2+ solvation ., How we characterize the ion-solvation layer from our simulated trajectories is described in section S3 in the S1 Text ( Fig s3 in S1 Text ) ., We have further classified the outer sphere Mg2+ present in the ion-solvation layer into two categories based on their number of associated phosphate groups:, ( i ) Single phosphate coordinating Mg2+ ( Fig 4b ) , which efficiently neutralize the negative charge of the adjacent phosphate ( Fig 4d ) , and, ( ii ) multiple phosphate coordinating Mg2+ ( Fig 4c ) ., The key role in stabilizing the structure is played by such Mg2+ bridging multiple phosphates , which can act as glue in compact structures by holding a number of negatively charged phosphates together in close proximity ., The population shift coincident with multiple coordinated Mg2+ ions with increasing Mg2+ concentration directly supports their role in stabilizing the structure ( Fig 4e ) ., We have also investigated the thermodynamic impact of Mg2+-mediated phosphate contacts ( PHOSCont: total number of pair-wise phosphate-phosphate contacts ) on the energy landscape as a function of overall folding progress , expressed by the number of native contacts ( NCont ) , as shown in Fig 4f–4h ., As we increase the Mg2+ concentration the broad minimum that appeared around NCont~800 , involving partially folded open conformations , gradually becomes more stabilized ., Concurrent enrichment of phosphate-phosphate contacts extends the contour of the minimum asymmetrically toward higher PHOSCont ., Additionally , by 8 mM Mg2+ , the bound-like pre-organized state grows with substantial population , stabilized again by phosphate connections ( Fig 4h ) ., We analyzed long equilibrium trajectories of the apo- and bound-forms of SAM-II slightly below the folding temperature in order to capture the essential characteristics of the pre-organized state , and also to compare this state with the fully folded ligand bound state ., We have evaluated the distribution of native contact formation in each segment of secondary structure as a function of the total Q at different Mg2+ concentrations ., Plots show contact formation in P2b ( Fig 5a–5d ) and the triplex interaction between helix P2b and loop L1 ( Fig 5e–5h ) , which are most affected by Mg2+ concentration ., Data for the nonlocal contacts of P1 , L3-P1 , P2a , which appear only marginally affected by Mg2+ concentration , are shown in Fig s4 in S1 Text ., The two distinct basins visible at low Mg2+ , for the P2b helix and L1-P2b triplex contacts correspond to the pre-organized ( at higher Q ) and open states ( at lower Q ) ., At increasing Mg2+ , the populations gradually shift towards the pre-organized state ., Around 8 mM Mg2+ , the dominant contribution arising from this pre-organized triplex to the conformational space is evident from Fig 5c and 5g ., Ligand binding also strongly favors structure formation , even at moderate Mg2+ , as the ligand bridges the gap between L1 strand and P2b helix , producing the fully formed triplex ., Motivated by our 13C-CEST profiles for SAM-II and their Mg2+ dependence we have explored the free energy landscape of the SAM-II riboswitch using a recently developed all-atom SBM that includes explicit Mg2+ ions , Debye-Hückel treatment of implicit KCl interactions , and the effects of Manning condensation to accurately account for the ion atmosphere around the RNA ., Our results support a mechanism involving Mg2+ induced pre-organization followed by conformational selection by the ligand , SAM , as we speculated in an early study 47 ., The free energy analysis validates the observations of that pre-organization , providing an atomistic and thermodynamic basis for the enhanced population of a partially collapsed , pre-organized ensemble at sufficiently high Mg2+ concentration in the absence of ligand ., We observe three distinct sets of conformations in the folding free energy landscape of ligand-free SAM-II riboswitch:, ( i ) an ensemble of unfolded conformations ,, ( ii ) a broad ensemble of partially folded open conformations , and, ( iii ) an ensemble of pre-organized bound-like conformations ., As we increase magnesium concentration beyond 2–4 mM , the bound-like ensemble is further stabilized , shifting the equilibrium toward the pre-organized states ., All the experimental results from our 13C-CEST profile , recent SAXS , single molecule FRET , and size-exclusion chromatographic studies are assembled and found to be in good agreement with the present simulation results ( Fig 2 ) ., At higher concentrations , Mg2+ stabilizes compact structures by coordinating multiple charged phosphate groups of RNA in close proximity ., The experimental results , together with free energy landscapes confirm that sufficient Mg2+ can indeed promote stable ligand binding in the SAM-II riboswitch , and is likely the structural basis for the switching control of protein translation ., While this structural pre-organization of SAM-II can assist in rapid ligand recognition , our study suggests that a sufficiently high concentration of Mg2+ is necessary to capture those pre-organized states ., Only when the system achieves a well-organized ion solvation layer at high Mg2+ , the effect of additional Mg2+ seems limited ., This layer involves a number of Mg2+ ions , each coordinating with multiple phosphate groups ., Mg2+ ions thus serve as glue to the negatively charged phosphates and facilitate the structural compaction ., We note that while chelated Mg2+ may play an important role in other riboswitch RNAs , no specific chelated ions have been reported so far in the SAM-II system ., Our molecular simulation trajectories also allow us to pinpoint the structural basis of the effect , revealing that triplex interaction between the helix P2b and its association with the L1 strand dominate the process of pre-organization as summarized in Fig 5a–5c and 5e–5g , showing the gain of structure with increasing Mg2+ ., In the final step , ligand binding firmly bridges the extended gap between L1 and P2b , which seems otherwise not achievable through the addition of small , dynamic Mg2+ alone ., But although P2b and its connection with L1 can be secured by the ligand , its presence again alone cannot fully stabilize the overall structure without addition of significant amount of Mg2+ ( Fig 2h ) ., These findings suggest that a sufficiently high concentration of Mg2+ is necessary to stabilize the pre-organized triplex and then the presence ligand promotes the native triplex formation , as summarized in Fig 5d and 5h ., We note that triplexes have recently emerged as important players in gene regulation by non-coding RNAs 50–53 ., Base triples also play a role in RNase P and the Diels-Alder ribozyme 54 ., Heroic calculations , as such recent microsecond explicit solvent simulations of riboswitches , will also shed light on these effects , especially regarding the role of solvation 55 ., Nucleic acid-ion interactions make a substantial energetic contribution in the stabilization of the native state of RNAs , including complex formation with proteins and other macromolecules 56 ., The dynamics of nucleic acids are also found to be strongly influenced by the motion of their ion atmospheres ., Relative to other ionic species , Mg2+ can efficiently support a close assembly of negatively charged phosphates by mediating favorable interactions among them ., Other earth alkali metals/divalent ions ( e . g . Ca2+ ) and even monovalent ions are also able to induce similar transitions , albeit at higher concentration ., Our early SEC elution profiles for SAM-II show that the transition midpoint in presence of Potassium ( K+ ) alone occurs only at K1/2 ≈ 25 mM ., The midpoint for Calcium ( Ca2+ ) is Ca1/2 ≈ 8 mM , compared to 6 mM for the Mg2+ ion 40 ., This is a direct result of the larger charge/radius ratio of magnesium 40 , 57 ., Thus , having these special characteristics , Mg2+ efficiently helps pre-organize the system and enables access to the partially collapsed states that are further stabilized by ligand binding ., The general importance of Mg2+ for the stability of compact RNA structures supports a possibly universal role of conformational selection in ligand-binding RNAs , such as riboswitches , aptamers , and possibly protein-binding RNAs ., A detailed thermodynamic understanding of the underlying landscape will indeed enable greater control of riboswitch regulation , highly sought after by researchers in synthetic biology who are currently employing riboswitches as ligand-dependent ‘knobs’ to control desired gene expression 58 ., Our all-atom structure-based model ( SBM ) has proven successful in describing the dynamics of numerous proteins and macromolecular complexes 59–63 ., To elucidate RNA free energy landscapes under the influence of Mg2+ , models capable of quantitatively describing the ion atmosphere are needed , including ionic condensation around the negatively charged phosphate groups of RNA ., Early studies have simply included electrostatic effects in SBM of RNA via repulsive Debye-Hückel interactions , thus treating all ions implicitly 29 , 30 ., Recently , our group developed a more detailed model of RNA electrostatics and applied it within all-atom structure-based molecular dynamics simulations ., Our model treats Mg2+ ions explicitly to account for ion-ion correlations neglected by mean-field theories 38 ., The KCl buffer , which completes the experimental setup , is treated implicitly by a generalized Manning counter ion condensation model 38 , 64 , since mean-field theories correctly assess the charge densities of monovalent K+ and Cl- ions ., Classical Manning counter-ion condensation theory was originally developed for understanding the low concentration limiting behavior of polyelectrolyte chains modeled with an infinite line of charge ., Folded RNA , however , is not a line of charge ., To account for the compact and irregular structures of RNAs and the effects of varying ion concentrations , we improve the Manning counter ion condensation model to handle electrostatic heterogeneity , making the condensed charge density a dynamical function of each phosphate coordinate ., KCl screening is characterized by a Debye-Hückel potential ., Removal of the continuum screening ions from the inaccessible volume of RNA is a substantial extension to Manning counter-ion condensation ., The model has been tested against experimental measurements of excess Mg2+ associated with RNA , characterizing the Mg2+-RNA interaction free energy ., This hybrid SBM has opened up new possibilities to study various structural and functional processes of RNA that are essentially controlled by ions 38 ., In the present study we used this recently developed all-atom hybrid SBM to understand the conformational transition of SAM-II and the corresponding Mg2+ sensitivity ., The energy function used in this model is given below ,, Φ=ΦSBM+ΦMg-Size+Φion-effect, ( 1 ), where , ΦSBM is the all-atom SBM potential ensuring a global minimum in the landscape for the native state of RNA ., The SBM potential is composed of two general types of interactions:, ΦSBM=Φlocal+Φnon-local, ( 2 ), where , Φlocal characterizes the local interactions that encode covalent bonds and torsional angles , maintaining the correct local geometry and chirality ., Φnon-local comprises two non-local contributions:, ( i ) an attractive term that is applied specifically to all tertiary interactions determined from the native structure ,, ( ii ) the general repulsive interactions , that describe the excluded volume by symmetric hard potentials ( to avoid any unwanted chain crossing ) ., ΦMg-Size adds the excluded volume interactions involving the explicit Mg2+ ions , regulating RNA-Mg2+ and Mg2+-Mg2+ interactions ., Φion-effect accounts for all interactions between charges in the system which consist of the fixed charge distribution of the RNA and the dynamic contribution from the ions ., Mg2+ and phosphate charges interact via a Debye-Hückel potential with a screening term that depends , in turn , on the distribution of the monovalent ions ., The monovalent ions , K+ and Cl- from the added salt , fall into two categories: screening ions and Manning condensed ions ., The screening ion density is obtained using Debye-Hückel electrostatics ., The density of the Manning-condensed ions is modeled as the sum of two normalized Gaussian distributions where the center of each Gaussian is located on the position of the negatively charged phosphate group ., All the condensation variables along with the explicit Mg2+ and RNA coordinates are evolved with Langevin dynamics 38 ., The mathematical formulations of all the terms and the related parameterizations are discussed in depth in section S4 in the S1 Text ., The umbrella sampling method 65 was used to sample the conformational space of SAM-II riboswitch along the reaction coordinate , Q , which is the fraction of intra-molecular native contacts in the riboswitch ., The Weighted Histogram Analysis Method 66 was then used to calculate the thermodynamic quantity , G ( Q ) ., The detail is described in section S5 in the S1 Text ., CEST data were collected using a pseudo-3D HSQC experiment with the B1 field offsets ( -600 to 600 Hz ) incremented in an interleaved manner with 3 references ( no CEST period ) 47 ., A total of 1024x16 complex points were recorded 40 with 32 transients with a recovery delay of 1 . 5 s for a total experimental time of approximately 12 hr for each spin-lock field ., A CEST saturation period of 100 ms was used for the base and 200 ms for ribose ., The pulse program used was an adaptation of a previously published one without the need for selective pulses 47 ., We used a two-state model to fit each of the three profiles of the selectively labeled carbon ( ribose C1’ and base C6 ) and quantitatively extracted the carbon chemical shift ( Δω ) , the exchange rate , and the population of the minor state based on the Bloch−McConnell 7x7 matrix 47 ., The CEST data was plotted as I ( t ) /I ( 0 ) versus spin-lock offset ( Hz ) and was fit by numerically solving the matrix exponential for the CES

Introduction, Results, Discussion, Methods

Our 13C- and 1H-chemical exchange saturation transfer ( CEST ) experiments previously revealed a dynamic exchange between partially closed and open conformations of the SAM-II riboswitch in the absence of ligand ., Here , all-atom structure-based molecular simulations , with the electrostatic effects of Manning counter-ion condensation and explicit magnesium ions are employed to calculate the folding free energy landscape of the SAM-II riboswitch ., We use this analysis to predict that magnesium ions remodel the landscape , shifting the equilibrium away from the extended , partially unfolded state towards a compact , pre-organized conformation that resembles the ligand-bound state ., Our CEST and SAXS experiments , at different magnesium ion concentrations , quantitatively confirm our simulation results , demonstrating that magnesium ions induce collapse and pre-organization ., Agreement between theory and experiment bolsters microscopic interpretation of our simulations , which shows that triplex formation between helix P2b and loop L1 is highly sensitive to magnesium and plays a key role in pre-organization ., Pre-organization of the SAM-II riboswitch allows rapid detection of ligand with high selectivity , which is important for biological function .

The presence of positively charged metal ions is essential to maintain the structural fold and function of RNA ., Among different metal ions , magnesium is particularly important for the stability of RNA because it can efficiently support a close assembly of negatively charged phosphate groups in an RNA fold ., The SAM-II riboswitch is an example of a classical pseudoknot fold , which binds S-adenosyl methionine , stabilizing an alternate folded form to inhibit gene expression ., In our early 13C- and 1H-chemical exchange saturation transfer ( CEST ) experiments , we found a conformational transition between a minor , partially closed and a major , open state conformation in the absence of ligand ., Our CEST experiments at different magnesium concentrations now suggest that magnesium ions can induce a conformational pre-organization in the apo SAM-II riboswitch , which is expected to facilitate ligand binding ., To understand the microscopic details of this magnesium-induced transition , we perform all-atom structure-based molecular simulations including electrostatics and explicit magnesium ions ., Our free energy calculations reveal that the partially closed pre-organized state is further stabilized with increasing magnesium concentration ., This is in excellent agreement with our 13C-CEST profile , SAXS , and size-exclusion chromatographic data , and with recent single molecule FRET experiments ., Our results suggest that a sufficiently high concentration of magnesium is essential to pre-organize the apo SAM-II riboswitch .

chemical compounds, phosphates, fluorophotometry, elution, thermodynamics, research and analysis methods, separation processes, spectrum analysis techniques, rna structure, fluorescence resonance energy transfer, magnesium, chemistry, molecular biology, spectrophotometry, free energy, physics, biochemistry, rna, biochemical simulations, rna folding, nucleic acids, biology and life sciences, physical sciences, computational biology, chemical elements, macromolecular structure analysis

journal.pcbi.1004923

Stochastic Simulation of Biomolecular Networks in Dynamic Environments

Dynamic simulation is an essential and widespread approach for studying biomolecular networks in cell biology 1 ., However , the computational resources required can quickly become limiting for several reasons ., Cellular networks are complex , containing many biomolecular species and reactions ., The effects of biochemical stochasticity can be pervasive at the single-cell level 2 , 3 , implying that stochastic simulation approaches are often needed ., And cells do not live in isolation , which requires simulation on multiple scales , ranging from the single cell to large ensembles of communicating cells 4 , 5 ., In these circumstances , parsimonious models of intracellular networks offer dimension reduction 6–8 and significant advantages 9 ., However , such models often only provide accurate descriptions when they include the effects of interactions with other fluctuating processes in the cell and of signals arising extracellularly 10–12 ., While it is straightforward to write a Chemical Master Equation describing the stochastic dynamics of these models , it is usually impenetrable to analysis and one needs to make use of simulation methods ., The stochastic simulation algorithm ( SSA ) 13 , 14 allows only the random timing of reactions in the network model to be taken into account ( often known as intrinsic noise ) , but cannot be used when other processes interacting with the network cause its propensities to fluctuate between reaction occurrences ., The SSA assumes constant propensities between reactions ( and hence exponentially distributed waiting times ) ., Here we present a new approach relaxing this assumption , called Extrande , for stochastic simulation of a biomolecular network of interest embedded in the dynamic , fluctuating environment of the cell and its surroundings ., An extensible implementation of Extrande for general reaction networks with multiple inputs is given in the S1 File ., Biological processes that interact with the network or model of interest are sometimes called extrinsic processes 15 ., They often significantly change the stochastic behaviour and dynamics of the network 16 , 17 ., We briefly give two illustrations of the biological importance of extrinsic processes as motivation for the development of our approach , the first well-established , and the second considered here ., First , although intrinsic noise is an important contributor , extrinsic processes are known to be a substantial and sometimes dominant source of variation in gene expression levels across cells and over time 18–21 ., We are now beginning to understand the underlying biological sources 22 , which include effects related to circadian oscillations , temperature , chromatin remodelling , the cell-cycle and pulsatile transcription factors 23 , 24 ., To understand gene expression , it is therefore essential to move beyond the SSA , which can only account for intrinsic noise , and to include other sources of variation ., Second , fluctuations in the expression , degradation and recycling of proteins inevitably affect the way networks containing those proteins function and the extent of stochasticity in the input they provide to other networks ., Fluctuations in the component proteins of signal transduction networks limit information transfer 25 , affect transduction network ‘design’ 26 and , although often overlooked , are inevitably conveyed ( as extrinsic inputs ) to the networks regulated by signaling ., Here , the computational advantages of Extrande will allow us to demonstrate how fluctuations in the protein componentry of signal transduction networks are conveyed to signaling outputs and place strong constraints on the design of networks determining cell fate , thus influencing the distribution of phenotypes at the population level ., Without the ability to simulate biomolecular networks that are exposed to fluctuating inputs , the ability to address such questions is severely restricted ., There are two existing approaches to stochastic simulation of reaction networks subject to dynamic , fluctuating inputs ., The first class of algorithms 5 , 13 , 27 implements the SSA , under the approximation that the input remains constant between the occurrences of any two reactions ., However , this approximation can give spurious results even when dynamic inputs to the network are changing relatively slowly ., We term these collectively the Slow Input Approximation method ( SIA ) ., The second class of algorithms 28–30 involves step-wise numerical integration of reaction propensities until a target value for the integral is reached ., Algorithms in this class would be ( conditionally ) exact , if it were not for the presence of numerical error in integration , but can impose large and impractical computational burdens , especially when cell ensembles are studied ., We term these collectively the integral method ( distinguishing next and direct integral approaches below ) ., We perform a comparative analysis of both methods with Extrande and demonstrate that our method offers an accurate and computationally efficient alternative approach ., Extrande involves no analytical or numerical integration but instead relies on ‘thinning’ techniques 31 , 32 ., Other approaches using rejection methods have also recently been proposed as a means to tackle systems with time-dependent propensities 12 , 33 ., The stochastic simulation algorithm ( SSA ) 13 , 14 allows simulation of biomolecular reaction networks taking into account the discreteness of these systems as well as the intrinsic randomness in the timing of reaction events ., The SSA assumes that the propensity of each reaction channel to fire , hence the probability of the reaction to occur over a small time interval , remains constant between reaction events ., This naturally restrains the use of SSA to simulate networks embedded in dynamic , fluctuating environments because the reaction propensities then become time-varying quantities under the influence of extrinsic processes ., Extrande ( Box 1 ) —or Extra Reaction Algorithm for Networks in Dynamic Environments—allows exact stochastic simulation of any downstream reaction network , conditional upon a time course of the dynamic inputs that is simulated up-front ., The method involves no analytical or numerical integration , though we give a connection to the direct integral method below , and instead makes use of point process ‘thinning’ techniques 31 , 32 , where some simulated events are discarded ., The only error incurred is any error associated with the input pre-simulation , typically an approximate simulation of a stochastic differential equation ( Box 1 ) ., The Extrande approach can be understood as introducing an extra , ‘virtual’ reaction channel into the system ( whose occurrence does not change molecule numbers ) ., The propensity of the extra channel is designed to fluctuate over time so that ( when added to the sum of all other reaction propensities ) the total propensity in the augmented system becomes constant between events and equal to an upper bound on the sum of the propensities in the original system ., To accomplish this , the method exploits the exogeneity of the dynamic inputs—the assumption of negligible retroactivity 35 from network to inputs ., In particular , their exogeneity means that Extrande is able to make use of the ‘future’ trajectory of the inputs to find an upper bound , B , on the total propensity , which is valid over a certain time interval L ( see Step 3 , Box 1 ) ., Simulation of the augmented system is feasible by means of an SSA-like algorithm ., The method uses the bound on the total propensity to generate a putative reaction reaction time τ ( Step 4 ) ., If the reaction time exceeds the time horizon L , it is rejected; the system time advances by L ( Step 6 ) , and the procedure restarts by determining a new bound ., Otherwise , time advances by τ and a reaction is chosen based on the updated reaction propensities ( at time t+τ ) ( Steps 8–15 ) ., The reaction events of the virtual channel are discarded , leaving those of the other channels—because the simulated timing and types of the biochemical reaction channels are unaffected by the behaviour of the extra channel , the result is a trajectory of the original system ( see Methods ) ., We study the decision to enter competence ( for uptake of extracellular DNA ) by the model organism Bacillus subtilis ., It is well established 39–41 that the source of differentiation of 10–20% of the cell population under stress conditions is fluctuations in transcription of the master competence regulator , ComK ., The ComS-MecA-ComK competence module is regulated by the activated transcription factor pComA , the output of the transduction mechanism relaying extracellular , quorum sensing signals ( CSF and ComX ) , see Fig 3A ., We study the effect of this upstream signaling on differentiation into the competent phenotype ., A useful approach to understanding the structure-function relationship in systems biology is to rewire networks found in nature and compare function with the wild-type , which can then shed light on why apparently similar network structures were not adopted naturally 42 ., In the wild-type , upstream signaling acts via activation of the ComS promoter by pComA binding ( Fig 3A , thick black arrow ) ., We compare the behaviour of wild-type cells to those with a Synthetic Decision-Making network ( SynDM ) which is regulated , in addition , via activation of the ComK promoter by pComA binding ( red dashed arrow ) ., We model ComK-driven progress and entry into functional competence , and write Progress ( t ) = k ∫ 0 t ComK ( s ) d s , where k is an effective rate of ComK-driven differentiation ., A cell is taken to enter ( functional ) competence at the time when Progress ( t ) = 1 ., The value of the parameter k is set so that the wild-type and SynDM networks give equal fractions of competent cells with a constant level of pComA ( 1000 molecules ) ., We tune rate parameters associated with the ComK promoter of the SynDM network so that the fraction of SynDM cells entering competence ( 0 . 18 ) is the same as for wild-type cells , in the absence of fluctuations in pComA levels ( see S1 Text ) ., A table listing all reactions and parameter values used in our models of the competence module of wild-type B . subtilis and the SynDM networks is given in the S1 Text ., We use the linear noise approximation ( LNA ) 43 to model the the upstream signaling ( with CSF and ComX fixed at steady-state levels ) , giving a mean for pComA of 1000 molecules throughout ., Importantly , we include in the model gene expression and degradation of the proteins comprising the signal transduction mechanism because it is now understood that the resultant variation has important effects on signaling and information transfer 26 ., A single Ornstein-Uhlenbeck ( OU ) process is sufficient to closely match the mean , variance and autocorrelation function of pComA given by the LNA ( see S1 Text ) ., We therefore use a single OU process for the pComA input in what follows ., A range of protein lifetimes is considered , consistent with the broad range of cell-cycle periods observed for bacteria under different growth conditions 44 , where nutrient limitation can result in periods in excess of 10h ., Our baseline LNA model of the upstream signaling module gives a lifetime and CV of pComA fluctuations equal to 5h and 0 . 35 ., We take the pComA input to be exogenous to the ComS-MecA-ComK competence module since it is in high abundance relative to the 2 promoters it binds ( the only interaction between the two modules ) ., The importance in determining cell fate of the time taken for the cell to complete different differentiation programs ( to the point of irreversible commitment ) has recently been emphasised 45 ., The SynDM network creates a differentiated sub-population by activating the differentiation program in most or all of the cell population ( Fig 3C & 3D ) , with entry to competence the outcome of a ‘race’ to differentiate over the relevant time window ., In the SynDM network , binding of pComA to the ComK promoter results more often in periods of non-zero ComK expression than in the wild-type population , but when such periods occur , they are less sustained ( see Fig 3B–3D , and Fig . E in S1 Text ) ., The typical rate of progress of a SynDM cell to competence is increased by a higher level of pComA ( see Fig . E in S1 Text ) , and extrinsic fluctuations in the pComA level therefore affect the fraction of cells entering competence ( Fig 3C & 3D ) ., In contrast , the wild-type activates the differentiation program in a smaller sub-population , the size of which is under modest regulation by pComA ( Fig 3F ) —a high proportion of the active wild-type cells then enter competence because , once activated , ComK expression rarely deactivates in the wild-type ( see Fig 3B , and Fig . E in S1 Text ) ., We find two important advantages of the wild-type design ( in addition to the implied reduction in the metabolic cost of gene expression ) ., First , the fraction of cells entering competence is considerably more robust to the fluctuations from upstream signaling in pComA ( Fig 3E ) ., For example , with the baseline model of upstream signaling , the SynDM network has a competent fraction ( 40% ) which is more than 2 . 25 times the competent fraction when pComA is held constant at its mean level , whereas the competent fraction of wild-type cells ( 17% cf 18% ) has changed very little ., The difference in robustness is explained by the sensitivity of the probability of competence for a SynDM cell as a function of the time average of the signal , 〈pComA〉 , which switches quite rapidly from zero to one ( Fig 3F ) ., Since the fraction of competent cells is equal to the average of ProbCompetence|〈pComA〉 over the distribution of 〈pComA〉 ( which is approximately the distribution of pComA for longer lifetimes ) , the competent fraction increases in the presence of extrinsic fluctuations for SynDM ( recall the mean of pComA is 1000 molecules ) ., In contrast , ProbCompetence|〈pComA〉 is approximately linear for the wild-type network , which implies that the competent fraction depends largely on the mean of pComA alone ., Such plots ( Fig 3F ) should prove a useful diagnostic tool for the design of synthetic decision-making networks ., The second advantage of the wild-type design is that the fraction of cells entering competence is also considerably more robust than SynDM to heterogeneity across the cell population in the rate at which ComK-driven differentation proceeds ( Fig 3G ) ., The reason is evident from the progress to competence trajectories in Fig 3B–3D ., We note that fluctuations from upstream signaling in pComA can also cause decreases in the fraction of competent SynDM cells , as seen for higher rates of differentiation ( Fig 3G ) ., Heterogeneity in the rate at which differentiation programs proceed is inevitable where cellular decisions are executed by large gene expression networks and involve substantial physiological changes 46 ., These in silico experiments ( Fig 3 ) , made computationally feasible by Extrande , cast light on the wild-type network design in which quorum signaling input to the competence decision-making network ( ComS-MecA-ComK ) by the transcription factor pComA exerts its effect only at the promoter of ComS and not at the promoter of ComK ., The experiments reveal exquisite robustness of the wild-type design to fluctuations from upstream signaling and to heterogeneity in downstream processes , and demonstrate the computational potential of Extrande for in silico network design ., Stochastic simulation of biomolecular networks is now indispensable for studying biological systems , from small reaction networks to large ensembles of cells ., The effects of stochasticity can be pervasive at the single-cell level , determining the distribution of phenotypes in a population and thus potentially affecting evolutionary outcomes ., However , studying such phenomena requires stochastic simulation of a large ensemble of cells that can take into account both intrinsic and extrinsic sources of cellular variation ., This can be hugely costly in terms of CPU time , placing important in silico experiments out of reach ., Here we provide the new Extrande approach—for stochastic simulation of a biomolecular network embedded in the dynamic environment of the cell and its surroundings—which substantially increases the computational feasibility of such experiments without compromising accuracy ., We show that previous approaches to this problem either can fail dramatically , even when inputs vary relatively slowly , or impose impractical computational burdens due to costly numerical integration of reaction propensities ., Given a simulated trajectory of fluctuating network inputs , the Extrande approach provides a conditionally exact solution that can speed up simulation by several orders of magnitude compared to integral methods ., In practice , we find that integral methods suffer from the high cost of propensity evaluations during numerical integration ., Extrande bypasses numerical integration by introducing an extra reaction channel—one designed to keep the total propensity of the ‘augmented’ system constant between events—hence making the problem of finding the time to the next event analytically tractable ., Importantly , our numerical results demonstrate that the overhead costs induced by the Extrande method—for example , due to thinning and rejection events , and due to obtaining the ceiling of the input process when a global ceiling is not available–are significantly lower than the cost of accurate numerical integration ., In practice , we observe speed-ups by a factor as great as 2 . 5×104 ( Fig 2C ) ., Recent work 12 proposes to handle fluctuating environments in a different manner , by deriving a network model for the biochemistry that takes account of the dynamic input and follows the correct ( marginal ) probability law ., Explicit simulation of the input is bypassed ., The resultant ‘uncoupled’ network model has time-varying reaction propensities and can then be simulated using integral or thinning methods ., However , analytical derivation of the uncoupled network model is not always possible , particularly when there are multiple inputs ., The accuracy of the method then depends on finding suitable approximation schemes ., There are two main limitations of modelling using the Extrande method ., The first is that Extrande , being a method of obtaining trajectories of the chemical master equation ( with time-dependent propensities ) , has the same applicability limitations as the master equation; namely there is an implicit assumption that the system is dilute ( point particles ) and well-mixed , conditions which are not met when molecular crowding is significant 47 , 48 ., The second limitation is that Extrande assumes that the inputs influence the system of interest but the latter does not influence the inputs ( which implies the inputs can be pre-simulated ) ., Hence the method is useful , for example , to understand how certain external stimuli such as light and temperature can affect the stochastic dynamics of a system ., For the case of a chemical stimulus , the method can provide an accurate description of the stochastic dynamics if the system and its output do not significantly feedback to adjust the original chemical stimulus , for example by a regulatory mechanism ., We exploit the benefits of the proposed Extrande simulation method here to study the decision-making behaviour of a quorum sensing population of bacterial cells ., The in silico experiments presented ( Fig 3 ) took approximately two computing months using Extrande ( and an Intel Xeon , 3 . 3GHz quad-core processor with 32GB of RAM ) , but would have been prohibitive using the integral method due to the approximate 70-fold slow down needed to ensure even modest accuracy ( see Fig . D in S1 Text ) ., The results elucidate the costs and benefits of alternative network designs for the probabilistic differentiation of a sub-population of cells in response to upstream signaling ., Our findings argue for the biological significance of fluctuations in signaling inputs that arise from synthesis and degradation of the protein componentry of signal transduction networks , and show that these fluctuations have important consequences for downstream networks such as those deciding cell fate ., We expect the accuracy and reductions in CPU time made possible by Extrande to help open up the landscape of computationally feasible simulation of biomolecular networks and cell ensembles ., Extrande thus has the potential to accelerate both understanding of molecular systems biology and the design of synthetic networks ., The Extrande approach relies on augmenting the reaction network with an extra , ‘virtual’ channel ( giving the augmented system , Z ) , so as to make simulation of the augmented system feasible , while ensuring that the simulated timings and types of biochemical reactions are unaffected by the firings of the extra channel ., In the Extrande method , the conditional propensity of the extra channel depends on the history of the extra channel ( as well as on the history of the original system , H t X ) , and so does the upper bound ., A related Proposition in 32 does not allow for this dependence ( see S1 Text ) ., We therefore provide the new proof below ., To see the dependence on the extra channel , note that the bound is in general updated in Step 3 of the Extrande algorithm ( Box 1 ) after each firing of the extra channel ., The reaction network to be simulated ( Box 1 ) has the number of molecules of each species at time t given by, X ( t ) = X ( 0 ) + S R ( t ) ,, where R, ( t ) = {R1, ( t ) , … , RM, ( t ) } is the vector of processes counting the number of times each biochemical reaction channel fires during the time interval 0 , t , and S = {v1 , … , vM} is the stoichiometric matrix ., The ‘Poisson’ or random time-change representation 49 expresses R, ( t ) in terms of M independent , unit rate Poisson processes , Y, ( t ) = {Y1, ( t ) , … , YM, ( t ) } , and so can be written here as, X ( t ) = X ( 0 ) + S Y 1 ∫ 0 t a 1 X ( s ) , I ( s ) d s , . . . , Y M ∫ 0 t a M X ( s ) , I ( s ) d s T , ( 1 ), where I is the possibly multivariate input , superscript T denotes transpose of a vector , and ajX, ( s ) , I, ( s ) is the propensity of the jth reaction , for j = 1 , … , M , conditional on { H s X , I } ., We denote by I ( the σ-field generated by ) the entire trajectory of the input ., We introduce as a simulation device the extra , virtual reaction RM+1: ∅ → ∅ , to form the augmented system Z ( t ) = X ( t ) R M + 1 ( t ) = X ( 0 ) 0 + S 0 0 1 R ( t ) R M + 1 ( t ) ., The random time-change representation of the augmented system is in terms of ( M+1 ) independent , unit rate Poisson processes , Y, ( t ) = {Y1, ( t ) , … , YM+1, ( t ) }, Z ( t ) = Z ( 0 ) + ( S 0 0 1 ) × ( … , Y j ( ∫ 0 t a j X ( s ) , I ( s ) d s ) , … , Y M + 1 ( ∫ 0 t a M + 1 ( s ) d s ) ) T ( 2 ), where aM+1, ( s ) is the propensity of the extra reaction channel ( conditional on { H s Z , I } ) , and where we set ajX, ( s ) , I, ( s ) , for j = 1 , … , M , as the propensity of the jth reaction conditional on { H s Z , I } , which now includes the history of the extra channel , RM+1 ., Notice that Eq 2 is identical to Eq 1 in its expression of the original system , X, ( t ) , or equivalently of R, ( t ) ., Therefore , if the propensity aM+1 is chosen to somehow make simulation of R, ( t ) , RM+1, ( t ) straightforward , we are able to simulate our target , R, ( t ) , by simulating the augmented system in Eq 2 and then ignoring RM+1, ( t ) ., To do this , let B, ( t ) be an ( H t Z , I ) -measurable random variable satisfying ( with probability 1 ) that, a 0 ( t ) = ∑ j = 1 M a j X ( t ) , I ( t ) ≤ B ( t ) , t ≥ 0 ,, so that B, ( t ) is a stochastic upper bound for the total biochemical reaction propensity ., Now define the propensity of the extra channel ( conditional on { H t Z , I } ) as:, a M + 1 ( t ) = B ( t ) - a 0 ( t ) ., The ground process ( see S1 Text ) of R, ( t ) , RM+1, ( t ) has propensity ( conditional on { H t Z , I } ) given by ∑ j = 1 M + 1 a j ( t ) = B ( t ) , by construction ., The Extrande method chooses the stochastic bound , B, ( t ) , so that it is constant between firings of the augmented system ( see Box 1 ) , which makes straightforward the simulation of the ground process of R, ( t ) , RM+1, ( t ) ., We write the ith occurrence time of the ground process of R, ( t ) , RM+1, ( t ) as Ti , i = 1 , 2 , … It is now the case that, Prob { T i + 1 - T i ≤ t | T 1 , Z 1 , . . . , T i , Z i , I } = 1 - exp { - t B ( T i ) } ,, where Zi is the channel corresponding to the ith firing ., The waiting time has an exponential distribution and the occurrence times {T1 , T2 , …} are therefore just those of a ( H t Z , I ) -Poisson process with propensity B, ( t ) , and can be simulated analogously to the SSA as in Step 4 of Box 1 ., What remains is to assign each firing time Ti to one of the ( M+1 ) channels of the augmented system ., We do the allocation sequentially , using the result from counting process theory 50 that , for j = 1 , … , ( M+1 ) :, Prob { Z i + 1 = j | T 1 , Z 1 , . . . , T i , Z i , T i + 1 , I } = a j X ˜ ( T i + 1 ) , I ˜ ( T i + 1 ) B ( T i ) , ( 3 ), where we have used the left-continuous versions ( X ˜ ( t ) , I ˜ ( t ) ) of ( X, ( t ) , I, ( t ) ) , and B ˜ ( T i + 1 ) = B ( T i ) ., Eq 3 is implemented by Steps 9–15 in Box 1 ., The intuition for Eq 3 uses Bayes’ theorem ., Consider a small interval of time dt ., The probability that the channel is the jth one given that some reaction fires at time Ti+1 , since the probability of more than one reaction can be neglected , is given by d t · a j ( X ˜ T i + 1 , I ˜ T i + 1 ) / d t · k = 1 M + 1 ∑ a k ( X ˜ T i + 1 , I ˜ T i + 1 ) ., The target of the Extrande simulation , R, ( t ) , is now obtained by ignoring all the firing times of the extra channel after simulation of the augmented system is complete ., This completes the proof ., ■ We note that the condition limt → ∞ Rj, ( t ) = ∞ ( j = 1 , … , M ) is needed for the representation in Eq 1 , but is not needed for the validity of the Extrande method ., The random time-change representation is used here to make the proof more accessible ., The Extrande algorithm results in a probability law , P , under which the functions ajX, ( t ) , I, ( t ) give the propensities of the biochemical reactions conditional upon ( H t Z , I ) ., Because the ajX, ( t ) , I, ( t ) are ( H t X , I ) -measurable , they also give the ( H t X , I ) -conditional propensities of the biochemical reactions under P , as required of the probability measure P resulting from the Extrande algorithm ., Finally , we remark that a description equivalent to the random time-change representation , Eq 1 , is the Chemical Master Equation 49 ., Specifically , for the conditional probability P ( n , t ) = Prob ( X ( t ) = n | X ( 0 ) = n 0 ; I ) one can write, d P ( n , t ) d t = ∑ j = 1 M a j n - v j , I ( t ) P ( n - v j , t ) - a j n , I ( t ) P ( n , t ) , ( 4 ), whose propensities are time-varying , stochastic functions due to the dependence on the input process .

Introduction, Results, Discussion, Methods

Simulation of biomolecular networks is now indispensable for studying biological systems , from small reaction networks to large ensembles of cells ., Here we present a novel approach for stochastic simulation of networks embedded in the dynamic environment of the cell and its surroundings ., We thus sample trajectories of the stochastic process described by the chemical master equation with time-varying propensities ., A comparative analysis shows that existing approaches can either fail dramatically , or else can impose impractical computational burdens due to numerical integration of reaction propensities , especially when cell ensembles are studied ., Here we introduce the Extrande method which , given a simulated time course of dynamic network inputs , provides a conditionally exact and several orders-of-magnitude faster simulation solution ., The new approach makes it feasible to demonstrate—using decision-making by a large population of quorum sensing bacteria—that robustness to fluctuations from upstream signaling places strong constraints on the design of networks determining cell fate ., Our approach has the potential to significantly advance both understanding of molecular systems biology and design of synthetic circuits .

Simulation algorithms have become indispensable tools in modern quantitative biology , providing deep insight into many biochemical systems , including gene regulatory networks ., However , current stochastic simulation approaches handle the effects of fluctuating extracellular signals and upstream processes poorly , either failing to give qualitatively reliable predictions or being very inefficient computationally ., Here we introduce the Extrande method , a novel approach for simulation of biomolecular networks embedded in the dynamic environment of the cell and its surroundings ., The method is accurate and computationally efficient , and hence fills an important gap in the field of stochastic simulation ., In particular , we employ it to study a bacterial decision-making network and demonstrate that robustness to fluctuations from upstream signaling places strong constraints on the design of networks determining cell fate .

signaling networks, cell differentiation, circadian oscillators, simulation and modeling, developmental biology, mathematics, network analysis, chronobiology, research and analysis methods, computer and information sciences, gene expression, biophysics, physics, biochemistry, biochemical simulations, genetics, biology and life sciences, physical sciences, computational biology, numerical integration, numerical analysis, biophysical simulations

journal.pgen.1001302

Pervasive Adaptive Protein Evolution Apparent in Diversity Patterns around Amino Acid Substitutions in Drosophila simulans

A central challenge of evolutionary biology is to elucidate the nature of adaptive changes to the genome: do they comprise a negligible or substantial fraction of differences among species ?, When they occur , are they driven by strong positive selection or are they fine-tunings of minor consequence to fitness ?, In Drosophila , perhaps the most studied taxon in these respects , there are conflicting accounts regarding the intensity of selection driving adaptations 1–4 but accumulating lines of evidence suggest that adaptation may be prevalent 5–7 ., The evidence is based primarily on two kinds of signatures that beneficial substitutions leave in their wake ., The first is an excess of divergence at functional sites compared to that expected under neutrality , detected using the approach introduced by McDonald and Kreitman 8–11 ., Numerous studies based on extensions of this approach indicate that approximately one in two amino acid and one in five non-coding differences between Drosophila species may be adaptive 7 , 11–14 ., These findings remain tentative , however , because other factors , and notably plausible demographic scenarios , could cause a substantial overestimation of the fraction of beneficial substitutions 7 , 8 , 15–17 ., Moreover , McDonald-Kreitman based approaches can provide only very limited information about the strength of positive selection ., The second footprint of adaptation is in diversity patterns ., When a rare or new allele is favored and fixes in the population , it drags closely linked neutral alleles to loss or fixation ., This “selective sweep” leads to a transient reduction in levels of neutral diversity around a beneficial substitution , where the size of the affected region decreases with the recombination rate and increases with the intensity of positive selection 18–20 ., In accordance with a model of recurrent selective sweeps , levels of synonymous diversity across the genomes of a number of Drosophila species increase with rates of crossing over 21–23 and decrease with increasing numbers of amino acid substitutions 2 , 3 ., Making reliable inferences about adaptation based on these relationships has been challenging , with two decades of effort focused on distinguishing the effects of positive selection from those of background ( i . e . , purifying ) selection and from possible mutagenic effects of recombination 5 , 24–29 ., By necessity , previous studies relied on limited summaries of the data , thereby losing much of the information carried by the spatial signature of beneficial fixations ., In particular , measurements of diversity , recombination , and functional divergence were taken in arbitrarily chosen window sizes , making it harder to distinguish the effects of adaptation from other evolutionary forces 29 , 30 , and likely biasing estimates of adaptive parameters of interest ( e . g . , the rate and intensity of selection ) 7 ., As an illustration , based on the relationship between diversity levels and amino acid divergence seen in 100 kb windows , Macpherson et al . 3 inferred few beneficial amino acid substitutions with a large selective coefficient of ∼1%; in contrast , focusing on the same relationship in individual genes , Andolfatto 2 inferred many beneficial amino substitutions with a selective coefficient of ∼10−3%; the two studies differed in other regards , but the disparate conclusions may reflect in part the choice of window size 7 ., In summary , despite accumulating evidence that adaptation may be widespread in Drosophila , we still lack characterizations that capture genome-wide signatures that are specific to adaptive evolution and do not rely on an a priori choice of scale ., Here , we take advantage of genome-wide variation data from Drosophila in order to produce a readily interpretable characterization of the effects of positive selection that overcomes a number of limitations ., To do so , we consider the average level of neutral diversity as a function of distance from amino acid substitutions ., Our reasoning is as follows: Beneficial amino acids that fixed in the recent evolutionary past ( ∼Ne generations 20 ) should create a trough in diversity levels around them , whereas amino acid substitutions that were selectively neutral or occurred farther in the past should have little effect on diversity patterns ., If we consider the effects of all amino acid substitutions in the genome jointly , and a non-negligible fraction of amino acid fixations were favored – as McDonald-Kreitman based estimates suggest – then we should expect a trough in the average level of neutral diversity around amino acid substitutions ., The depth of this trough is expected to increase with the fraction of beneficial amino acid substitutions , and its width will reflect the intensity of selection driving these substitutions ., In contrast to previous approaches , this characterization does not depend on an a priori choice of window size , and captures much more of the footprint of adaptive substitutions ., To generate this plot , we use autosomal amino acid substitutions on the lineage leading from the common ancestor of Drosophila simulans and D . melanogaster to D . simulans , relying on the genomes of D . erecta and D . yakuba as outgroups 31 ., As a measure of neutral diversity , we consider the number of synonymous polymorphisms divided by the overall number of codons at a given distance from an amino acid substitution ., The polymorphism levels in D . simulans are measured using a recent dataset of six inbred lines 5 , down-sampled to have a uniform sample size of 4 lines at ∼50% of the codons in the genome ., Ideally , we would like to plot diversity levels as a function of genetic distance from amino acid substitutions , since the expected reduction in diversity depends on genetic rather than physical distance from the selected loci ., Since there are no high-resolution estimates of recombination rates in D . simulans , we use physical distance instead , but consider only regions for which the homologous regions in D . melanogaster have an estimated recombination rate above 0 . 75cM/Mb ., The collated plot in Figure 1A ( red ) thus obtained is averaged over n\u200a=\u200a26 , 834 amino acid substitutions ., Because the plot is constructed by conditioning on a substitution at the center , diversity patterns could be distorted even in the absence of adaptive evolution ., Namely , if mutation rates vary across the genome then they might , on average , be elevated near substitutions ., Considering the average synonymous divergence between D . melanogaster and D . yakuba as a proxy for the mutation rate confirms this expectation , as it reveals a small increase near substitutions ( Figure 1B ) ., To correct for this elevation in rates , we divide the average level of diversity around amino acid substitutions at a given distance by the average divergence ( Figure 1C ) ., Moreover , as a control , we compare the patterns around amino acid substitutions with plots that were constructed analogously but around synonymous substitutions instead ( Figure 1A–1C: black ) 28 ., As predicted by a model of recurrent selective sweeps , we find a clear reduction in diversity levels around amino acid substitutions relative to the synonymous control ., This reduction is statistically significant within a window of ∼15kb around amino acid substitutions ( at the 1% level , as assessed by bootstrapping; see Text S1 ) ., Farther from substitutions , where sweeps are unlikely to have an effect on diversity , the curves for synonymous and amino acid substitutions are indistinguishable ., This pattern is robust to the effects of synonymous codon usage bias ( Figure 4 in Text S1 ) , as well as to changes in the recombination rate threshold ( Figure 5 in Text S1 ) , and to the choice of outgroup used to correct for the mutation rate ( not shown ) ., In addition , we see similar patterns when we examine the substitutions that occur on any one of the autosomal chromosome arms ( Figure 6 in Text S1 ) ., This pattern is a distinctive signature of adaptive evolution ., Demographic processes would not lead to systematically decreased diversity around amino acid substitutions ., In turn , for background selection to generate the observed trough centered on amino acid substitutions , its effects in regions of the genome with moderate to high recombination rates would have to be strong enough to lead to both a substantial reduction in diversity and to the fixation of many weakly deleterious amino acid mutations ., Modeling indicates that , given plausible parameters for Drosophila , this is highly unlikely 32 ., Our analyses also reveal that amino acid substitutions are clustered near one another ( Figure 2A: red ) ., This clustering is greater and more localized than the clustering of synonymous substitutions around amino acid substitutions ( Figure 2A: black ) , implying that it is caused by more than the spatial distribution of exons in the genome and an elevated mutation rate near amino acid substitutions ., The difference between the clustering of amino acid and synonymous substitutions further suggests that variation in constraint and possibly in adaptability among and within genes contribute to the pattern for amino acid substitutions ( 33; also see Text S1 ) ., Aside from being an interesting finding in itself , this clustering could influence the observed reduction in diversity ., If two amino acid substitutions occur in close proximity and one led to a recent selective sweep , the reduction in diversity that it caused will also be observed around the other substitution ., This effect will reduce diversity around both non-synonymous and synonymous substitutions , but it will have a larger effect around amino acid substitutions because the density of amino acid substitutions nearby is on average greater ( Figure 2A ) ., Indeed , the level of synonymous diversity decreases strongly with the density of amino acid substitutions surrounding a substitution ( Figure 2B; Figure 8 in Text S1; Spearmans ρ\u200a=\u200a−0 . 93 for amino acid substitutions and ρ\u200a=\u200a−0 . 88 for synonymous substitutions; p<10−15 for both ) , consistent with previous studies 2 , 3 ., We also find , however , that the average level of synonymous diversity around amino acid substitutions is consistently lower than that around synonymous substitutions when the two are matched for the density of amino acid substitutions in their vicinity ( Figure 2B; Figure 8 in Text S1; signs test p<10−4 ) ., In other words , there is a substantial relative reduction in diversity around amino acid substitutions that is not explained by the amplifying effects of clustering ., In addition to providing compelling evidence for the prevalence of beneficial amino acid substitutions , the collated plot carries information about selection parameters , as the shape of the trough in diversity is indicative of the rate of adaptive protein evolution and of the distribution of selective effects of fixations ., To learn about these parameters , we develop a coalescent-based model for average diversity levels as a function of distance from an amino acid substitution , accounting for their clustering ( see Text S1 ) ., Using this model , we infer adaptive parameters by jointly maximizing the composite-likelihood of diversity patterns as a function of different distances from the focal substitution ( i . e . , the likelihood of points along the entire curve ) , thus mining a richer summary of the data than previous approaches ., When we assume that a fraction α of beneficial substitutions were driven by a selection coefficient s and the rest were neutral , we estimate that ∼5% of the substitutions were beneficial with a relatively strong selection coefficient of ∼0 . 4% ( Table 5 in Text S1 ) ., Using a Gamma distribution for the selection coefficients , α increases to ∼6 . 5% and the average selection coefficient remains similarly high; despite the additional parameter , the likelihood is barely higher ( Table 5 in Text S1 ) ., These estimates are relatively insensitive to assumptions about other parameters ( with the exception of the assumptions about recombination rates , as discussed below ) ; in particular , simulations suggest that the estimated strength of selection is robust to demographic assumptions ( see Text S1 for details ) ., A visual comparison suggests a reasonable fit of these models to the data ( Figure 3A ) ., However , the inference based on models with one selection coefficient , or even a Gamma distribution of coefficients , might be dominated by the broad features of the plot , such that any narrower trough caused by beneficial substitutions with weaker selection coefficients could be overlooked ., A closer look around the focal substitutions supports this notion , revealing a small trough inside the main trough , on the scale of several hundred bps , which is not captured by either of the two models ( Figure 3B ) ., We therefore consider another model , with two beneficial selection coefficients ., Using it , we estimate that ∼13% of the substitutions were beneficial , ∼3% with a large selective advantage of ∼0 . 5% and the rest with a much weaker effect , of approximately one hundredth of a percent ( Table 5 in Text S1 ) ., A mixture model with two exponentials reveals a similar picture: ∼4% of substitutions are estimated to come from a distribution with a mean selective coefficient of ∼0 . 5% and 11% from a distribution with a mean of ∼4·10−5 ( Table 5 in Text S1 ) ., Importantly , both models provide a substantially better fit to the data ( Table 5 in Text S1 ) and they capture the smaller as well as the larger troughs in diversity ( Figure 3A and 3B ) ., In turn , estimates under a model with three beneficial selective coefficients are similar to those obtained in model with only two and offer no improvement to the fit ( Table 5 in Text S1 ) ., Taken together , these findings indicate that selective sweeps are driven by two classes of beneficial fixations: a minority with large beneficial effects that account for most of the reduction in diversity and a majority with much weaker effects ., Moreover , they help explain why previous inferences based on the signatures of sweeps in Drosophila yielded markedly different estimates ( ranging over three orders of magnitudes ) 1–4 ., Our estimates of the fraction of beneficial amino acid substitutions ( ∼13% ) are on the same order of magnitude but lower than previous McDonald-Kreitman based estimates ( ∼50%; cf . 7 ) ., Some of this difference might arise from violations of the assumptions on which the inferences rely; in particular , in our approach , that adaptive parameters have remained constant in the D . simulans lineage , or in McDonald-Krietman based inferences , that the efficacy of purifying selection has not changed markedly 8 , 16 , 34 ., An intriguing alternative is that the two approaches are actually estimating parameters of somewhat different modes of adaptation ., Our inference is based on the effects of beneficial substitutions that arise from new mutations and likely misses some contribution of adaptation from standing variation ., Specifically , a subset of beneficial substitutions could stem from previously neutral or deleterious alleles that were segregating in the population before a change in the environment rendered them beneficial ., If these alleles were young when the environment changed , they would still generate the signature of a selective sweep and contribute , at least partially , to our estimated fraction of beneficial substitutions ., This is likely for alleles that were previously deleterious and at mutation-selection balance , but also possible for neutral alleles 35–37 ., If , however , the segregating alleles were older when they became beneficial and at higher frequency in the population , they would lead to a negligible effect on diversity and would therefore not contribute to the signature on which our inference relies ., These beneficial substitutions would nonetheless contribute to an excess of non-synonymous divergence compared to the neutral expectation , and should therefore be picked by the McDonald-Kreitman based inferences , leading to higher estimates of adaptive substitutions than obtained by our approach ., Other modes of adaptation , such as polygenic selection , may also contribute differentially to the two inference methodologies 38 ., We note that a current limitation of our inference is its reliance on rough estimates of the recombination rate , and its assumption of a constant rate per base ., In the logistic approximation to the trajectory of a beneficial allele , the expected reduction in diversity as a function of distance from the beneficial substitution depends on s/r , where s is the selection coefficient and r is the genetic distance to the substitution ( Equation 2 in Text S1 ) ., This implies , for example , that if our inference relies on a recombination rate consistently two-fold greater than the real rate , our estimated selection coefficient will be two-fold overestimated ( see Table 3 in Text S1 ) ., We therefore consider our estimates of selection coefficients to be rough approximations ., In addition , heterogeneity in the recombination rate , such as is known to exist in other taxa ( e . g . , 39 , 40 ) , could also affect our inferences ., The heterogeneity would have to be of a highly specific nature in order to account for our finding of two markedly different scales of selection coefficients , but at the moment , we cannot rule out the possibility ., For these reasons , it would be important to revisit the inference once we possess high-resolution genetic maps in D . simulans ., In summary , our findings establish a distinctive , genome-wide signature of adaptation in D . simulans , suggesting that many amino acid substitutions are beneficial and are driven by two classes of selective effects ., Enabled by a richer summary of diversity patterns that avoids an a priori choice of scale , these conclusions offer a coherent interpretation of the results of previous inferences ., It will now be interesting to see whether similar findings emerge in other Drosophila species , which vary in their recombination rates , effective population sizes , and ecology ., We reconstructed the sequence of the ancestor of D . melanogaster and D . simulans in order to identify substitutions along the D . simulans lineage ., For that purpose , we use a four species alignement from the 12 Drosophila genomes project 31 consisting of D . simulans , D . melanogaster , D . yakuba and D . erecta , and removed codons containing gaps in either of them ., We then inferred the ancestral sequences using PAML , with the CODEML model and the ( ( D . mel , D . sim ) , ( D . yak , D . ere ) ) tree 41 ., To measure polymorphism levels at coding regions of the D . simulans genome , we used resequencing data from six inbred lines of D . simulans and their alignment with D . melanogaster 5 ., We applied quality control filters and randomly down-sampled the remaining codons to four , in order to maintain a uniform sample size in measuring polymorphism ., In the end , we retained ∼50% of all protein-coding DNA ., Unless otherwise noted , our analysis was performed on data from autosomal regions , for which the sex-averaged recombination rate in the homologous region of D . melanogaster was greater than 0 . 75cM/Mb ( using the genetic map as in 3 ) ., See Section 1 in Text S1 for more details ., We used synonymous polymorphisms to measure the average levels of diversity as a function of distance from amino acid and synonymous substitutions along the D . simulans lineage ., To measure the average level of diversity at distance x , we divided the number of codons segregating for a synonymous polymorphism by the overall number of codons observed in the D . simulans polymorphism dataset at distance x from one of the amino acid ( or synonymous ) substitution ., In order to control for variation in the neutral mutation rate around substitutions , we calculated the average synonymous divergence around both amino acid and synonymous substitutions ., For that purpose , we identified synonymous substitutions between D . melanogaster and D . yakuba and measured the average level of divergence at distance x by dividing the number of codons exhibiting a synonymous substitution between D . melanogaster and D . yakuba by the overall number of codons observed in the alignment of these species at distance x from one of the amino acid ( or synonymous ) substitutions ., For further details and the robustness analysis , see Sections 2–4 in Text S1 ., The shape of the collated plot around amino acid substitutions carries information about the rate of adaptive protein evolution and the intensity of selection driving it , two parameters of long-standing interest ., To learn about these parameters , we developed a model describing the expected neutral diversity levels around substitutions , which relies on Gillespies pseudohitchhiking coalescent model 42 ., We then used a composite likelihood approach 43 to estimate the parameters ., For a description of the approach and assessments of its reliability , see Section 6 in Text S1 .

Introduction, Results/Discussion, Materials and Methods

In Drosophila , multiple lines of evidence converge in suggesting that beneficial substitutions to the genome may be common ., All suffer from confounding factors , however , such that the interpretation of the evidence—in particular , conclusions about the rate and strength of beneficial substitutions—remains tentative ., Here , we use genome-wide polymorphism data in D . simulans and sequenced genomes of its close relatives to construct a readily interpretable characterization of the effects of positive selection: the shape of average neutral diversity around amino acid substitutions ., As expected under recurrent selective sweeps , we find a trough in diversity levels around amino acid but not around synonymous substitutions , a distinctive pattern that is not expected under alternative models ., This characterization is richer than previous approaches , which relied on limited summaries of the data ( e . g . , the slope of a scatter plot ) , and relates to underlying selection parameters in a straightforward way , allowing us to make more reliable inferences about the prevalence and strength of adaptation ., Specifically , we develop a coalescent-based model for the shape of the entire curve and use it to infer adaptive parameters by maximum likelihood ., Our inference suggests that ∼13% of amino acid substitutions cause selective sweeps ., Interestingly , it reveals two classes of beneficial fixations: a minority ( approximately 3% ) that appears to have had large selective effects and accounts for most of the reduction in diversity , and the remaining 10% , which seem to have had very weak selective effects ., These estimates therefore help to reconcile the apparent conflict among previously published estimates of the strength of selection ., More generally , our findings provide unequivocal evidence for strongly beneficial substitutions in Drosophila and illustrate how the rapidly accumulating genome-wide data can be leveraged to address enduring questions about the genetic basis of adaptation .

Characterizing the nature of beneficial changes to the genome is essential to our understanding of adaptation ., To do so , researchers identify and analyze footprints that beneficial changes leave in patterns of genetic variation within and between species ., In order to teach us about adaptive evolution , these footprints need to be specific to positive selection as well as rich enough to allow for reliable inferences ., Here , we identify such a footprint: a pronounced trough in the average levels of genetic diversity surrounding amino acid substitutions throughout the D . simulans genome ., Based on this pattern , we infer that approximately 13% of amino acid substitutions were beneficial , a minority of which ( 3% ) conferred a large selective advantage of nearly 0 . 5% and the majority of which ( 10% ) conferred a much smaller advantage of about 0 . 01% ., These findings offer insights into the distribution of selection effects driving beneficial changes to the D . simulans genome and suggest how the widely varying estimates obtained in previous studies of Drosophila may be reconciled ., Moreover , the approach that we introduce is readily applicable to other taxa and thus should help to gain important insights into how the rate and strength of adaptive evolution vary depending on life-history , population size , and ecology .

evolutionary biology, genetics and genomics

journal.pgen.1002275

Genome-Wide Association Study Identifies Four Loci Associated with Eruption of Permanent Teeth

Dental maturation is the process of exfoliation of primary teeth and eruption and calcification of permanent teeth that generally takes place between 6 and 13 years of age 1 ., One commonly used measure of dental maturity is the number of permanent teeth erupted at a given age 2 ., This is influenced by several factors , including gender 3–5 , malnutrition 6 , caries or trauma to the primary teeth 7 , 8 , ethnicity 9 and certain diseases 10–12 ., In addition , many dental traits are known to be substantially influenced by genetic factors 13–15 ., Although a recent study identified genetic variants for development of primary dentition in infancy 16 , genetic factors influencing the eruption of permanent teeth have not been identified ., To search for sequence variants associated with number of permanent teeth erupted , we carried out a GWAS in more than 5 , 100 women from the Danish National Birth Cohort ( DNBC ) 17 , who had records in the nationwide dental registry for children ( SCOR ) and replicated the findings in more than 3 , 700 individuals from Denmark and the US ( see Table S1 for description of study groups ) ., We analyzed the association between the number of permanent teeth erupted and 521 , 741 SNPs in 5 , 104 women with dental records from their childhood , and identified four loci strictly fulfilling genome-wide significance ( P<5×10−8 , see Figures S1 and S2 for quantile-quantile and Manhattan plots ) ., No other region showed a P-value<5×10−6 ., To confirm the observed associations we genotyped and tested the most significant SNP at each of the four loci in additional Danish samples , as well as tested for in silico replication in a US study of dental caries ., In the 3 , 762 additional individuals all SNPs replicated with P<10−4 and the combined P-values were <10−11 ( Table 1 ) ., Interestingly , two of the variants have been previously reported to affect primary tooth development 16 ., First , rs12424086 on chromosome 12q14 . 3 had a suggestive P-value ( 5×10−8<P<5×10−6 ) for number of primary teeth at age 1 ( P\u200a=\u200a3 . 64×10−6 ) 16 ., This SNP is about 120 kb downstream of HMGA2 ( see Figure 1 for genomic regions of the four associated SNPs ) and is also in linkage disequilibrium ( LD ) with rs1042725 ( r2\u200a=\u200a0 . 21 in HapMap Europeans , physical distance ∼6 kb ) , a SNP associated with adult and childhood height 18 ., Second , rs4491709 on chromosome 2q35 , is in LD with rs6435957 ( r2\u200a=\u200a0 . 73 in the DNBC I study group , physical distance ∼17 kb ) , a SNP that showed suggestive evidence for an association with number of primary teeth at age 1 ( P\u200a=\u200a3 . 64×10−7 ) 16 ., Furthermore , rs4491709 is in LD with rs13387042 ( r2\u200a=\u200a0 . 34 in the DNBC I study group ) , which is associated with breast cancer 19; the closest gene is TNP1 ( 160 kb telomeric ) ., For the genomic regions around the two other SNPs there were no previous significant GWAS results ., The third SNP , rs2281845 , on chromosome 1q32 . 1 is just upstream of CACNA1S ( voltage-dependent calcium channel , L type ) , a gene subject to mutation screens for malignant hyperthermia 20 , 21 and periodic paralysis 22 , 23 ., The LD block containing rs2281845 extends to TMEM9 ( transmembrane protein 9 , physical distance 22 kb ) ., The direction of the effect for rs2281845 in the US study group was - though not significant - opposite to the overall effect , which is most likely due to the relatively small sample size and greater variation in the phenotype ( individual values are only based on one observation ) ., The fourth SNP , rs7924176 , on chromosome 10q22 . 2 is intronic in ADK ( adenosine kinase ) , a gene that has been studied in the context of type 1 diabetes 24 , and is located in a broader region showing linkage with Alzheimers disease 25 ., We also looked at all variants reportedly associated with tooth eruption in primary dentition ( Table S2 ) ; among the 6 genome-wide significant SNPs , rs1956529 also showed a substantial effect on permanent tooth eruption , and three other loci had effects in the same direction with P-values between 0 . 01 and 0 . 17 ., The correlation between age at eruption of first tooth and permanent tooth eruption is evident ( r\u200a=\u200a−0 . 408 in a subset of 1 , 442 children from the DNBC with questionnaire data on age at first tooth eruption and SCOR data on number of permanent teeth erupted; later ages at first tooth eruption going along with lower age-adjusted standard deviation scores ( SDS ) for number of permanent teeth erupted ) ., We followed up all four SNPs associated with permanent tooth eruption in an on-going GWAS of primary dentition based on more than 6 , 000 individuals from the Avon Longitudinal Study of Parents and Children ( ALSPAC ) ( Table S3 and Text S1 ) ., Rs7924176 also reached genome-wide significance in the analyses for number of primary teeth erupted at age 15 months and time to eruption of first tooth ., The other three SNPs were at least nominally significant in the analysis of number of teeth erupted at age 15 months ( P between 0 . 028 and 3 . 5×10−6 ) , in all instances the same alleles associated with fewer teeth erupted in the respective dentition ., Given that both rs4491709 on chromosome 2q35 and rs1956529 on chromosome 14q24 . 1 , which associated with tooth eruption in primary dentition 16 , are close to breast cancer susceptibility loci , we checked all 19 SNPs associated with breast cancer in Caucasian study groups ( P<5×10−8 ) 19 , 26–32 ., Only two of the other 18 loci were nominally significant ( Table S4 ) without reaching a P<0 . 05/18 ., Thus , we could not find further evidence for a link to breast cancer ., Another trait of maturation , age at menarche , showed association with more than 30 SNPs in a recent GWA meta-analysis based on 87 , 802 women 33 ., We checked the SNPs reported for age at menarche in our permanent tooth eruption GWAS ( Table S5 ) and rs7821178 on chromosome 8q21 . 11 reached a P-value of 1×10−4 , which is significant after adjustment for 42 SNPs tested ., The allele increasing age at menarche also delayed tooth eruption , a pattern seen in all 6 SNPs with P<0 . 1 for permanent tooth eruption ., This could be due to some genetic variants regulating maturation in a more general way , even though the correlation between age at menarche and permanent tooth eruption is modest ( r\u200a=\u200a−0 . 095 in the combined DNBC I and DNBC II study groups , earlier age at menarche correlated with more permanent teeth erupted ) ., Results for the 180 SNPs recently reported to be associated with adult height 34 are given in Table S6 ., Apart from rs1351394 , which is in LD with the identified SNP rs12424086 in the HMGA2 region , rs6473015 ( P\u200a=\u200a6 . 1×10−6 ) was also significant after adjusting for the number of SNPs tested for height ., This SNP is on chromosome 8q21 . 11 within 90 kb of rs7821178 ( r2\u200a=\u200a0 . 78 ) , the age at menarche variant with low P-value for permanent tooth eruption ., The correlation between permanent tooth eruption and adult height is modest ( r\u200a=\u200a0 . 074 in the combined DNBC I and DNBC II study groups , more permanent teeth erupted correlated with increased adult height ) and there is no consistent direction among the height SNPs reaching nominal significance for permanent tooth eruption ., The correlation between age at menarche and adult height ( r\u200a=\u200a0 . 099 , earlier age at menarche correlated with decreased adult height ) is well known from the literature 35 , but is opposite to what would be expected just based on the correlation results for permanent tooth eruption ., Even though all correlations are modest , they underline that the interplay between these three growth and maturation traits is not straightforward ., To follow-up on the ( potential ) links to age at menarche , height and breast cancer , we contacted the latest published GWAS for these traits 26 , 33 , 34 and results for the reported four SNPs are displayed in Tables S7 , S8 , S9 ., For age at menarche there is no indication that the four SNPs play a role ( lowest P\u200a=\u200a0 . 15 ) based on results from the meta-analysis with 87 , 802 women of European descent ( Table S7 ) ., For height , results based on 133 , 000 individuals showed the expected signal for rs12424086 in the HMGA2 region , and rs4491709 had a modest effect that reached nominal significance ( P\u200a=\u200a0 . 02 , Table S8 ) ., Currently , several GWAS on breast cancer have been conducted , with the latest having a combined 2 , 839 cases and 3 , 507 controls at the initial GWAS stage 26 ., This group provided us with results for 1 , 693 cases vs . 5 , 588 controls ( Table S9 ) and rs4491709 ( which is in LD with the known breast cancer risk SNP rs6435957 ( r2\u200a=\u200a0 . 73 ) , reached nominal significance ( P\u200a=\u200a0 . 024 ) , for the three other regions the lowest P was 0 . 096 ., We investigated association with expression levels of nearby genes in an expression database for monocytes 36 ., The four SNPs were not genotyped directly , but several SNPs in LD with rs2281845 showed significant association with expression of TMEM9 , and the SNP with the strongest LD ( rs6667912 , r2\u200a=\u200a0 . 46 in HapMap Europeans ) had a P-value of 6×10−29 ., Similarly , we observed association of rs7924176 with expression of ADK via rs1874152 ( r2\u200a=\u200a0 . 71 in HapMap Europeans ) at a P-value of 2×10−42 ., The women from the DNBC provided a relatively homogeneous group with similar mean number of individual observations ., Therefore , we split the time period from age 6 to 14 years into four two-year periods to see the effect of the variants at different ages ( Figure 2 ) ., All variants showed significant effects on the number of teeth erupted in all four age periods , including the age group 12 to 14 years in which the permanent dentition was nearly or fully erupted ( mean number of erupted teeth is 26 . 5 ) ., The highest variation was observed in the age group 10 to 12 years ( standard deviation ( SD ) 4 . 7 teeth ) , and went along with the strongest per allele effect estimates ( −0 . 55 to −0 . 67 teeth ) ., Considering the combined number of delayed tooth eruption alleles , the 4 . 8% of children with 6 to 8 delayed eruption alleles had on average 18 . 5 ( SD 4 . 5 ) permanent teeth compared to 22 . 0 ( SD 4 . 2 ) in the 6 . 3% of children with 0 or 1 risk allele ( Figure 3 ) ., Effect estimates in the three other age categories ranged from −0 . 15 to −0 . 34 teeth per allele ., The variance explained by the four variants combined ranged from 1 . 5% ( age 12 to 14 ) to 3 . 0% ( age 10 to 12 ) ., Permanent tooth eruption happens earlier in girls than in boys , so we checked for gender differences in the replication groups , but there was no indication of heterogeneous effects between sexes ( results not shown ) ., However , the low number of male individuals in the study means that we only have limited power to detect effect differences between sexes ., We carried out the first GWAS for normal variation in the timing of permanent tooth eruption ., Using longitudinal dental data between age 6 and 14 years , we identified four loci with robust associations ., The main strengths of our study are:, i ) the comprehensive data set on dental exams during childhood , which allowed us to generate a mean SDS for the number of permanent teeth erupted ,, ii ) the Danish replication groups with comprehensive phenotype data from the same database as the initial study , and, iii ) the substantial sample size of more than 3 , 700 individuals at the replication stage ., The main limitations of the study are, i ) the lack of corresponding data on primary dentition , which is owed to the fact that there is usually no need to present children to the dentist in the time period where primary teeth erupt ,, ii ) the lack of other growth and maturation traits , because such data are not collected in connection with visits to the dentist and are not readily available otherwise ., The connection of our study to primary tooth eruption is obvious with rs7924176 reaching genome-wide significance in the analysis of number of permanent teeth erupted at age 15 months in the ALSPAC data , and the three other loci being at least nominally significant ., Looking at the 6 SNPs previously reported as genome-wide significant ( P<5×10−8 ) for primary dentition , rs1956529 in the RAD51L1 region showed association with permanent dentition ( p\u200a=\u200a3 . 6×10−5 ) and three SNPs had P-values between 0 . 01 and 0 . 17 , with the effect going in the same direction ., Thus , maybe all these SNPs are relevant in both tooth eruption periods , just with substantial variation in the strength of association between periods ., Children with data for both tooth eruption periods are necessary to see whether SNPs with strong effects on primary tooth eruption have an independent effect in permanent dentition ., On the other hand , there was no association with permanent tooth eruption for the two correlated SNPs in the EDA region ( both P>0 . 4 and effect in opposite direction ) , which means that there are also substantial differences between the genetic mechanisms driving primary and permanent tooth eruption ., The two identified SNPs previously reported in primary dentition are in LD with genome-wide significant SNPs in adult and childhood height ( chromosome 12q14 . 3 ) and breast cancer ( chromosome 2q35 ) ., Additional studies will be necessary to determine whether similar or different mechanisms explain the associations in these two regions ., We analyzed expression data for monocytes to investigate functional implications of the four SNPs ., Rs2281845 and rs7924176 could be linked to expression of TMEM9 and ADK , respectively , with rs7924176 being intronic in ADK ., However , the function of TMEM9 is unknown whereas ADK regulates the concentrations of extracellular adenosine and intracellular adenine nucleotides , providing no further insight into the potential role of these genes in tooth development ., Dental maturation was found to proceed rather independently of other forms of biological maturation in epidemiologic studies 1 , 37 and our comprehensive comparisons with other GWAS findings for adult height and age at menarche showed only modest overlap between associated variants ., Investigating permanent tooth eruption in 180 height loci and 42 age at menarche loci revealed three SNPs that were significant after adjustment for the number of variants tested for each trait , with the first signal being the height SNP in the previously discussed chromosome 12q14 . 3 region ., The two other SNPs are correlated and located on chromosome 8q close to PXMP3 ( peroxin 2 ) , with GWAS findings for both age at menarche and adult height , suggesting that this region is of general importance in growth and maturation ., Age at menarche is currently the only pubertal trait with GWAS findings , leaving the genetic background of pubertal growth and maturation poorly understood ., A focused investigation of the four SNPs in other puberty-related traits , e . g . skeletal maturation , Tanner score and growth spurt , could reveal whether these variants regulate maturation in a global way or are rather specific to dental maturation ., Basic characteristics of the study populations are provided as Table S1 ., For all study groups , all individuals with available information on permanent tooth eruption were analyzed , regardless of the phenotype they were recruited for ., Genotyping for the DNBC I and the US group was performed with Illumina ( Illumina , SanDiego , CA , USA ) Human 660W-Quad ( preterm birth ) and 610-Quad ( obesity and US ) bead chips ., Single SNP genotyping of rs2281845 , rs4491709 , rs7924176 and rs12424086 for the Danish replication samples was carried out at deCODE genetics using the Centaurus ( Nanogen , Bothell , WA , USA ) platform ., The initial GWAS was based on SNPs that passed quality control on both chips , SNPs were excluded based on a missing rate >2% , deviation from Hardy-Weinberg equilibrium ( P<10−4 ) or minor allele frequency <0 . 5% , individuals with more than 5% missing genotypes were also excluded ., The genotypes for the four selected SNPs in the replication stage did not show deviation from Hardy-Weinberg equilibrium or missing rates >2% ., Dental data for all Danish individuals were retrieved from the nationwide dental registry for children , SCOR , which was established in 1972 alongside the initiation of free municipal dental services to Danish children and adolescents from birth to the age of 18 years 49 ., Data were reported annually for all children until January 1st 1993 , from which date reporting was only mandatory for 5- , 7- , 12- , and 15-year-old children 50 ., All participants for the US sample underwent intraoral examinations by trained dental clinicians to collect phenotype data including number of erupted permanent teeth ., The study combined all observations between age 6 and 14 years ( starting with the 6th and stopping with the 14th birthday ) , the time period when eruption of permanent dentition usually occurs ., For each visit to the dentist the total number of permanent teeth ( excluding third molars ) was recorded , and regressed on age ., The resulting residuals were then standardized , and for each individual the mean residual across all available records was used as phenotype ., This approach was motivated by a recent study , which demonstrated that averaging over multiple records increases the power in quantitative trait analysis 51 ., For the DNBC I study group , we limited analyses to SNPs passing quality control on both chips and analyzed the two subgroups ( preterm birth study/obesity study ) together ., Initial GWAS analysis of mean standardized residuals was carried out by applying the Wald test under an additive genetic model ( also for chromosome X since all analyzed individuals are female ) as implemented in PLINK 52 ., We selected only SNPs with P<5×10−8 for replication , because SNPs in other regions did not show P-values<5×10−6 ., With the initial GWAS reaching genome-wide significance , the replication was basically of technical nature , aiming to confirm the results for the four SNPs with a second genotyping method ., The criterion for overall significance remained P<5×10−8 , but the additional study groups should not show great heterogeneity in terms of effect estimates and therefore an improvement of the combined P-values was expected ., P-values from the studies based on chip-typed individuals were corrected by applying genomic control ( estimated genomic inflation factors for the DNBC I and US study groups were 1 . 05 and 1 . 01 , respectively ) ., Combined effects and P-values were calculated using the inverse variance method as implemented in METAL 53 ., Though the number of permanent teeth was not normally distributed and the distribution of mean standardized residuals remained somewhat skewed , parametric tests were nevertheless carried out to determine the effect of the variants ., However , to protect against false positive findings due to the phenotype distribution , we additionally carried out a non-parametric analysis of our top SNPs from the four significant loci using Kruskal-Wallis tests in R and calculated combined P-values based on weighted Z-scores ( Table S10 ) ., Despite the reduced power of the non-parametric analysis ( which also does not account for the trend observed over the genotype groups ) , all SNPs reached genome-wide significance ( P<5×10−8 ) ., To control for possible population substructure , we performed multidimensional scaling analysis ( as implemented in PLINK 52 ) on the discovery data , using independent autosomal SNPs with missing call rates <1% , minor allele frequency >5% , and Hardy-Weinberg P-value>0 . 05 ., We utilized PLINKs LD pruning function to remove short and long-range LD ., The resulting 23 , 111 SNPs were analyzed along with founder genotypes from 11 HapMap phase III reference populations ., As expected most discovery samples clustered near the Caucasian populations from Utah and Tuscany , but 61 individuals fell more than 4 standard deviations away from the discovery set mean on one or more of the first five dimensions and were excluded from further analysis ., For all Danish replication sets , we obtained birthplace information from the Danish Civil Registry 54 , and only included individuals who were born in Scandinavia and whose parents were not born outside of Europe ., Data from the Danish Family Relations Database were used to exclude all individuals who were first or second degree relatives to an individual in the initial set or another subject already in the replication set ., The samples from the Denmark Roskilde and Glostrup groups were additionally genotyped on an ancestry informative microsatellite panel to confirm self-reported Danish European ethnicity , and individuals with <90% European ancestry were removed ., In the US study group we performed multidimensional scaling analysis to account for possible population substructure , resulting in the exclusion of 45 individuals ., Recruitment was partly family-based ( mainly sib ships ) , and we decided to keep one individual per pedigree based on the examination being closest to 10 . 0 years ( the midpoint of the investigated age interval ) , leading to the exclusion of 132 individuals ., In order to allow for comparisons with previous GWAS on related phenotypes , we separately imputed genotypes for the two GWAS included in the DNBC I study group applying MACH 55 ., The imputed genotypes were analyzed separately and meta-analyzed with METAL 53 ., ALSPAC http://www . bristol . ac . uk/alspac/; British 1958 Birth Cohort: http://www . b58cgene . sgul . ac . uk/; DNBC: http://dnbc . dk/; eQTL database: http://eqtl . uchicago . edu/cgi-bin/gbrowse/eqtl/; GENEVA: http://www . genevastudy . org; GIANT meta-analysis: http://www . broadinstitute . org/collaboration/giant/index . php/Main_Page; HapMap: http://hapmap . ncbi . nlm . nih . gov/; R: http://www . r-project . org/;

Introduction, Results, Discussion, Methods

The sequence and timing of permanent tooth eruption is thought to be highly heritable and can have important implications for the risk of malocclusion , crowding , and periodontal disease ., We conducted a genome-wide association study of number of permanent teeth erupted between age 6 and 14 years , analyzed as age-adjusted standard deviation score averaged over multiple time points , based on childhood records for 5 , 104 women from the Danish National Birth Cohort ., Four loci showed association at P<5×10−8 and were replicated in four independent study groups from the United States and Denmark with a total of 3 , 762 individuals; all combined P-values were below 10−11 ., Two loci agreed with previous findings in primary tooth eruption and were also known to influence height and breast cancer , respectively ., The two other loci pointed to genomic regions without any previous significant genome-wide association study results ., The intronic SNP rs7924176 in ADK could be linked to gene expression in monocytes ., The combined effect of the four genetic variants was most pronounced between age 10 and 12 years , where children with 6 to 8 delayed tooth eruption alleles had on average 3 . 5 ( 95% confidence interval: 2 . 9–4 . 1 ) fewer permanent teeth than children with 0 or 1 of these alleles .

While genome-wide association studies ( GWAS ) initially focused on the disease under investigation , additional findings in secondary traits have shown further benefits of having extensive phenotype data at hand ., Using records from the nationwide dental registry for children and genotype data from two GWAS , we were able to identify four genomic loci associated with permanent tooth eruption in children ., Two of the identified genomic regions had no previous GWAS findings , whereas two loci were reported in the context of primary dentition ., A follow-up in an on-going GWAS showed that rs7924176 also plays a substantial role in primary dentition ., During the age period of permanent tooth eruption many important developmental processes take place ., Thus , we suggest following up the four reported SNPs in other growth-related traits to further elucidate the genetic background of maturation .

genome-wide association studies, genome analysis tools, genomics, heredity, genetics, molecular genetics, biology, computational biology, genetics and genomics, human genetics

journal.pntd.0006059

TNF-α blockade suppresses pericystic inflammation following anthelmintic treatment in porcine neurocysticercosis

Neurocysticercosis ( NCC ) , an infection of the central nervous system ( CNS ) by the larval stage ( cysticercus ) of the parasitic cestode Taenia solium , is a major cause of epilepsy in developing countries and a serious public health burden 1–4 ., The disease is endemic to regions across the world where pigs are raised and allowed to roam freely with access to human waste 1 , 5 ., The occurrence of seizures and other symptoms of NCC depend on the number , location and distribution of cysticerci , the intensity of brain inflammation and the degenerative stage of the parasite , resulting in a wide variety of manifestations 2 , 6 ., A notable feature of T . solium infections is that viable cysts provoke minimal or no host-directed inflammatory responses ., However , degenerating cysts or cysts damaged by anthelmintic treatment provoke inflammatory responses that can have pathological consequences on brain tissues surrounding the dying parasite 2 , 5 , 7 ., Consequently , inflammation around degenerating cysts in the brain parenchyma generally results in seizures , whereas inflammation in the subarachnoid spaces causes diffuse and/or focal arachnoiditis frequently resulting in hydrocephalus , infarctions and nerve entrapments ., Cysts in the ventricles commonly cause hydrocephalus due to mechanical obstruction of cerebrospinal fluid ( CSF ) outflow or to ventriculitis and scarring 1 , 8 ., The pathological inflammatory response induced by cysticidal drugs can interfere with treatment ., Although corticosteroids are almost universally used to suppress inflammation and control symptoms , the ideal regimen for the safe and effective use of corticosteroids or other anti-inflammatory agents in multicystic or complicated NCC has not been determined ., As a result , the dose , duration and type of corticosteroid used are frequently based on the individual practitioner’s experience or preference 5 ., A better understanding of the acute inflammatory responses induced by treatment is necessary to formulate simple , safe and more effective treatment measures ., Studies of human and animal models of NCC indicate that inflammatory mediators produced by innate and adaptive immune cells play an important role in regulating inflammation both locally and systemically 9–16 ., We previously demonstrated that expression of mediators of inflammation such as tumor necrosis factor ( TNF ) -α , interleukin ( IL ) -6 and interferon ( IFN ) -γ was up regulated following anthelmintic treatment around cysts that displayed disruption of blood brain barrier integrity 17 ., These findings suggested points of attack to suppress specific pathways controlling treatment-induced inflammation to avoid the serious adverse effects of global immunosuppression associated with corticosteroids ., In the present study we focused on the TNF-α pathway of inflammation because of its importance in this infection ., Changes in expression of genes encoding a number of inflammatory mediators and regulatory factors following treatment with praziquantel were determined in pericystic brain tissue from infected pigs following blockade of TNF-α with etanercept ( ETN ) , a competitive inhibitor of TNF-α , and compared to corresponding tissues from a group of PZQ-treated pigs pretreated with corticosteroids and a control group of PZQ-treated pigs who did not receive any pretreatment ., Twenty-four T . solium-infected outbred pigs , confirmed by a positive tongue examination for cysts , were obtained in Huancayo , Peru , a town in a region of Peru endemic for cysticercosis ., Four healthy outbred uninfected pigs purchased in Lima , Peru served as a source of tissues to normalize the gene expression assays; they did not receive any treatment ., The four study groups included: untreated ( U ) , anthelmintic treatment with praziquantel ( PZQ , 100 mg/kg; P ) , dexamethasone ( DEX , and PZQ; DP ) and etanercept ( ETN and PZQ; EP ) ., The experimental design , including treatment and sample collection schedule is shown in Fig 1 ., Pigs were housed in the animal facility of the San Marcos Veterinary School ., A hundred and twenty hours after administration of PZQ , the pigs were anesthetized with ketamine ( 10 mg/kg , intramuscular injection ) and xylazine ( 2 mg/kg , both from Agrovetmarket SA , Peru ) , for an intravenous catheterization and infusion of Evans Blue ( EB ) and euthanized with sodium pentobarbital ( 25 mg/kg kg every 30 min for two hours , intravenous injection; Montana SA , Peru ) ., The study protocol and procedures were reviewed and approved by the Comité Institucional de Ética para el Uso de Animales–CIEA ( Institutional Ethics Commitee for the Use of Animals ) of the Veterinary School of San Marcos University in Lima , Peru ( Protocol numbers 006 for Universidad Nacional Mayor de San Marcos and 62392 for Universidad Peruana Cayetano Heredia ) ., The Comité is registered in the Office for the Wellbeing of Laboratory Animals of the Department of Health and Human Services of the National Institutes of Health with Policy Number A5146-01 ., All procedures used at the Veterinary Medicine Faculty of Universidad National Mayor de San Marcos ( FMV-UNMSM ) adhere to the International Guiding Principles for Biomedical Research that Imply the Use of Animals by the Council of International Organizations of Medical Sciences ( CIOMS ) , Geneva , 1985 ., As shown in Fig 1 , infected pigs were treated with a single dose of praziquantel ( PZQ , 100 mg/kg PO , on day 0; Montana SA , Peru ) , pretreated with DEX ( 0 . 2 mg/kg IM , on days -1 , +1 and +3; Química Farvet , Mexico ) , ETN ( 25 mg/pig on days -7 and day 0 , when PZQ treatment was also administered; Amgen , CA ) or no drugs and sacrificed 120h later ( n = 21 , total ) ., Three untreated infected pigs were used as controls ., Two hours before euthanasia , pigs were anesthetized and infused with 2% EB ( 80 mg/kg; Sigma-Aldrich , St . Louis , MO ) in saline solution ( NaCl 0 . 85% , Laboratories Baxter , Colombia , Baxter del Peru ) by intravenous injection via the carotid artery ., Just after euthanasia , the pigs were perfused with chilled saline solution containing heparin ( NaCl 0 . 85% with 10 U of heparin/mL ) and the brains were immediately removed to collect specimens ., Brains were sliced into 10-mm thick sections on dry ice ., The presence of unambiguously blue and clear stained cysts and tissues was documented by gross examination ., Tissues surrounding visible cysts ( pericystic “capsules” , histologically consisting of collagen and cellular infiltrate ) were sampled from each brain ( ~1 mL fragments ) and either placed in RNALater solution ( Invitrogen , Gaithersberg , MD ) for RNA extraction or fixed , together with the cyst , in 10% formalin ( PBS pH 7 . 2 with 3 . 7% formaldehyde ) for histological examination ., More cysts were selected for histology than for RNA studies because of relatively limited resources for the latter ( See S1 Table ) ., Paraffin sections were processed using standard procedures and stained with hematoxylin and eosin ( HE ) and Masson’s Trichrome stain ( MT ) ., Only capsules containing a whole cyst or a cyst wall surrounded by host tissue were selected for histological examination ., We focused on the blue cysts capsules for qPCR analysis for two main reasons ., Firstly , our previous studies had shown that significant upregulation of inflammatory and regulatory genes was apparent in blue capsules and not in clear capsules 17 , 18 ., Secondly , the proportion and number of clear capsules following PZQ treatment was dramatically reduced , resulting in unacceptably large variance in the parameters studied and making statistical inference unreliable or not possible ( See S2 Table ) ., Histological examination and analysis were performed exactly as reported previously 17 , 18 ., Briefly , a low power image of a section of whole cyst was first examined to assess the circumferential proportion associated with pericystic inflammation ., Higher power examination was then employed to determine the proportion of the cyst circumference showing each inflammatory stage ( IS ) present around each cyst ., The classification of the inflammatory stages followed the schema described by Álvarez et al 19 and Londoño et al 20 , and adapted by us in a previous study 18 ., According to this scheme , stage categorization of the inflammatory infiltrate was semi-quantitatively determined based on the average number of cells per high power field , and the thickness and location of type I collagen fibers around the cysts and in pericystic capsules 17 ., Using these measurements , IS 1 to 4 represented increasing severity of inflammatory reaction and pathology ., Similarly , cyst wall damage was categorized into four stages ( Damage Score: DS0-DS3 by severity of tissue disruption in the cysts , as outlined by Londoño et al . 20 ., Composite inflammatory ( IS ) and damage scores ( DS ) were determined for each cyst using the formula: Composite IS or DS score = Sum ( Score x % of cyst circumference ) x 100 ., The percentage of circumference was rounded off to increments of 25% ( i . e . , 0 , 25 , 50 , 75 or 100% ) ., IS and CD scores from the untreated ( U ) and PZQ-treated ( P ) groups pooled data from the experimental groups in the current experiments and with those from previous experiments of identical design published elsewhere 17 ., qPCR was performed on subsets of cysts selected from cysts with unambiguously clear or blue capsules ., The distribution of cysts from each experimental group is shown in S1 Table ., Fragments of brain tissue ( 50–100mg ) containing only cyst capsules ( no cyst ) were homogenized in 1 ml TRIzol ( Invitrogen , Gaithersberg , MD ) for standard RNA chloroform extraction ., cDNA was generated from 1 μg of total RNA using the High-Capacity cDNA Reverse Transcription Kit with multiscribe RT polymerase and random primers ( Applied Biosystems , Foster City , CA ) in 20-μl reactions by incubation for 10 min at 25°C followed by 60 min at 37°C , 5 min at 95°C on a thermocycler ( MJ Research PTRC-200 , BioRad-MJ Research , Hercules , CA ) ., Real time PCR ( qPCR ) was performed in 10-μl reaction volumes using the Taqman Gene Amplification System ( Applied Biosystems ) with commercially available primer-probe combinations using conditions recommended by the manufacturer ., We used 18S rRNA as a control gene to validate RNA integrity and primer probe pairs for porcine TNF-α , IL-6 , IFN-γ , IL-10 , CTLA4 , IL-13 , TGF-β , MMP1 , MMP9 , TIMP1 , TIMP2 , VEGF , PECAM1 , Ang1 and Ang2 genes ., qPCR reactions , run in triplicate , used the following cycling parameters: preincubation of 2 min at 50°C and 10 min at 95°C followed by 40 cycles of 15 sec at 95°C and 1 min at 60°C , on an AB StepPlusOne cycler ( Life Technologies , Grand Island , NY ) ., Results for each gene were expressed as relative to the expression of 18S rRNA using the X-fold value defined by the 2-ΔΔCT formula 21 ., The number of cysts analyzed for each marker differed within a given experimental group , because limitations on the amount of RNA extracted prevented us from analyzing each cyst for all the desired markers ., Non-parametric statistics , Mann-Whitney U test for two groups and Kruskal-Wallis test for multiple groups , were calculated using Prism® software ( Graphpad , San Diego , CA ) for comparisons of the above parameters between infected pigs that were not treated or treated with one of the two anti-inflammatory agents ( PZQ or ETN ) , and between clear cysts and those with EB staining ( blue cysts ) ., Corrections for multiple comparisons were applied for pairwise comparisons ., Differences with p-values of <0 . 05 were considered statistically significant ., Two-way contingency analysis , for example , comparing differences in proportions of cysts between PZQ-treated and PZQ plus DEX-pretreated pigs were performed by the Fisher’s exact test ., All infected animals had cysts scattered throughout the brain ., On gross examination , pericystic capsules were found either to be clear or stained blue due to extravasated EB ., As observed previously 22 , the proportion of capsules that had EB extravasation was increased 120h after PZQ treatment ( Fig 2A; p<0 . 001 , Fisher’s test ) ., The increase in proportion of blue capsules after PZQ treatment was not reduced by pretreatment with DEX or ETN ( Fig 2A; p>0 . 05 , Fisher’s test ) ., A total of 339 cysts were collected from all experimental animals , of which 233 cysts with intact pericystic capsules were examined histologically for both IS and DS ., To determine if pretreatment with DEX or ETN affected the degree of inflammatory infiltration around cysts or damage to cyst walls , we compared the IS and DS among the experimental groups ., No significant differences were noted in the IS or DS in clear capsules ( Fig 2B and 2C ) , likely due to the high variability in the scores ., However , IS and DS scores for blue capsules were higher in PZQ-treated pigs compared to untreated pigs ., Pretreatment with DEX or ETN reduced the IS of blue cysts compared to PZQ alone ( Fig 2B ) ., Analysis of cyst wall damage associated with PZQ treatment revealed that DEX pretreatment , but not pretreatment with ETN , resulted in a decrease in the DS ., ( Fig 2C ), These findings suggest dissociation between the regulation of inflammation by TNF-α , and the induction of cyst wall damage ., A smaller subset of cysts than those examined histologically was analyzed for expression levels of genes for molecules involved in tissue inflammation ( S1 Table ) ., Comparison of three markers of inflammation , TNF-α , IL-6 and IFN-γ , revealed that 120-h after PZQ treatment there was a significant up regulation of all three markers in blue capsules , as expected ( Fig 3A–3C , p<0 . 05 for each ) ., However , TNF-α blockade with ETN prior to PZQ treatment resulted in a profound decrease in the expression of TNF-α compared to PZQ alone ( Fig 3A; ~10-fold decrease to untreated baseline , p<0 . 0005 ) ., A smaller , but significant ( p<0 . 005 ) decrease was observed in DEX pretreated pigs ., IL-6 and IFN-γ were similarly inhibited in blue capsules by ETN pretreatment ( Fig 3B and 3C ) ., In clear capsules that lack the disruption of vascular integrity , gene expression of pretreatment with ETN or DEX could not be compared to PZQ alone ( see S1 Fig ) ., Remarkably , expression of TNF-α was lower in cysts from ETN pretreated pigs than from pigs that did not receive PZQ ( Fig 3A; p<0 . 05 , Mann-Whitney U test ) ., There was no significant corresponding inhibition of IFN-γ in the clear cysts ( S1 Fig ) ., We next analyzed the effect of ETN and DEX pretreatment on counter regulatory pathways ., In our previous study , we found variable inhibition of regulatory molecules , so that CD25 and CTLA4 transiently decreased at 48h post PZQ treatment , whereas IL-10 showed persistent decrease 48h post treatment and later 17 ., In the present study , IL-10 gene expression was inhibited from baseline after ( 120-h ) PZQ treatment , but neither DEX nor ETN reversed the PZQ-induced inhibition ( Fig 4A ) ., In contrast , the expression of three other regulatory molecules , CTLA4 , IL-13 and TGF-β , was significantly decreased by ETN ( but not by DEX ) pretreatment ( Fig 4B–4D ) ., While the inhibitory effect of ETN on gene expression of CTLA4 and TGF-β was only apparent around blue cysts , inhibition of IL-13 expression was also significantly decreased around clear cysts in pigs pretreated with ETN compared to DEX ( S1 Fig ) ., Unfortunately , there were insufficient numbers of clear cysts in the PZQ alone treated pigs for valid statistical inferences for comparisons with ETN or DEX pretreatment ., Genes associated with tissue remodeling and also in granuloma formation , such as matrix metalloprotease ( MMP ) 2 and MMP9 , as well as their regulators , the tissue inhibitors of metalloproteases ( TIMP ) 1 and TIMP2 , are up regulated following PZQ treatment in rodent models of NCC 23 , 24 and in infected pigs 17 ., Analysis of these genes in the present study revealed profound down regulation of MMP2 , MMP9 , TIMP1 and TIMP2 in blue cyst capsules from ETN-pretreated pigs compared to pigs who received PZQ alone ( Fig 5A–5D ) ., These observations are consistent with the global inhibition of gene expression that appears to accompany TNF-α blockade ., Interestingly , DEX pretreatment did not significantly down regulate transcripts that were markedly inhibited by the TNF-α blockade ., Expression of genes involved in endothelial activation such as PECAM1 , and angiogenesis such as VEGF , angiopoietin 1 and 2 ( Fig 5E–5H ) are commonly involved in the pathogenesis of parasitic infections 25–27 ., Our prior findings that showed increased vascular leakage in the cyst capsules following PZQ treatment of pigs 17 , 18 also suggest that endothelial integrity and function may be involved in the resulting inflammatory pathology ., We found that ETN pretreatment resulted in a significant inhibition of transcription of all these molecules , which were up regulated by PZQ treatment alone ., Taken together , these data reveal that TNF-α blockade resulted in transcription inhibition of a diverse range of inflammatory and regulatory pathway molecules and suggest an important role for TNF-α in regulating the inflammation that predictably follows PZQ treatment ., We also determined gene expression levels for the genes discussed above in tissues around cysts from the four experimental groups that did not have EB leakage and , therefore , were not demonstrating disruption of the BBB ( Fig 6 ) , referred to as “clear” cysts ., The total number of clear cysts was low because the PZQ treatment , which the three experimental groups received , appears to induce BBB leakage in the large majority of cysts ., Although there were decreases in gene expression of some inflammatory markers ( e . g . , TNF-α , IFN-γ ) in clear cysts from DEX and ETN-treated pigs that were similar to those seen in blue cysts , other inflammatory markers showed increased expression in DEX- and ETN-treated pigs ( e . g . , IFN-γ and Ang2; Fig 6C and 6G ) ., These differences between the blue and clear cysts probably reflect the lower level of tissue penetration of ETN in the latter due to an intact BBB , that may influence it’s effectiveness in the tissues ., In T . solium-infected humans and pigs , inflammation around dying cysts occurs frequently and predictably within a week of cysticidal treatment in humans and in animal models of NCC 28–33 ., In humans , the evidence for this manifest as an increased incidence of headaches , seizures and other neurologically based symptoms associated with the development of gadolinium enhancement and edema around some cysts seen on MRI imaging within the first week after initiation of cysticidal treatment ., Limited histopathological examination of brain tissues around degenerating cysts in patients post-treatment or with untreated “degenerated” cysts that shows infiltration of inflammatory cells 5 additionally supports the concept of treatment induced inflammation ., Previous studies performed by ourselves and others demonstrated similar post-treatment inflammatory reactions in pigs 17 , 18 , 22 , 33–35 ., In a practical sense , these detrimental side effects of antiparasitic treatment complicate medical treatment of NCC because they are a cause of morbidity and need to be prevented and controlled with corticosteroids , the use of which is associated with a variety of side effects 1 , 5 , 36 , 37 ., In a review focused on murine models of cysticercosis , one author discussed the concept of using approaches other than corticosteroids to inhibit inflammation associated with cysticidal treatment 38 ., In the CNS , TNF-α produced by migrated peripheral immune cells or by microglia and astrocytes in the presence of inflammation 39 , 40 plays an important role in inducing and maintaining inflammation that occurs in NCC ., Evidence for this includes high levels of TNF-α reported in CSF samples from patients 12 , 41–47 and increased expression of the TNF-α gene in pericystic capsules in pigs treated with PZQ 17 , 48 , 49 ., TNF-α has also been shown to be a key cytokine in maintaining inflammation in other inflammatory diseases involving the CNS , such as some causes of meningitis and autoimmune encephalitis 50–56 ., Therefore , we hypothesized that blockade of TNF-α should mitigate post-treatment inflammation in NCC ., Our data show that TNF-α blockade during PZQ treatment resulted in a broad and unique pattern of inhibition of gene expression for inflammation promoting proteins , regulating molecules , a number of pathways of tissue remodeling substances and molecules that modulate endothelial cellular function ., Notably , not all the genes tested were inhibited by ETN: for example , IL-10 expression in PZQ plus ETN treated pigs did not differ significantly from pigs receiving PZQ alone or PZQ plus DEX ., In contrast , DEX administration prior to PZQ treatment significantly inhibited only TNF-α , IL-6 , IFN-γ , TGF-β and Ang1 relative to PZQ treatment alone ( Figs 3 , 4 and 6 ) ., The lower anti-inflammatory responses to DEX compared to ETN was surprising , since corticosteroids are known for their potent and global immunosuppressive activity 57–59 ., However , pigs are known to be relatively insensitive to the immunosuppressive effects of corticosteroids 60 ., The effectiveness of ETN-mediated inhibition of multiple pathways of inflammation and tissue remodeling that is demonstrated by these data suggest a significant role for TNF-α in post-PZQ inflammatory responses An interesting finding in this study was an apparent dissociation between the effects of ETN on gene expression for inflammatory mediators and regulators ( Figs 3 and 4 ) and its effect on cellular infiltration in pericystic tissues ( Fig 2B ) ., TNF-α triggers a cascade of inflammatory cytokines , but also promotes endothelial cell contribution to local inflammation via the display of different combinations of adhesion molecules for leukocytes , including E-selectin , intercellular adhesion molecule-1 ( ICAM-1 ) and vascular cell adhesion molecule-1 ( VCAM-1 ) in a distinct temporal , spatial and anatomical pattern 61 , 62 ., In combination with the release of chemokines ( including IL-8 , MCP-1 and CCL2 ) 63 , these responses lead to recruitment of different populations of leukocytes , so blockade of TNF-α would normally be expected to inhibit cellular recruitment ., However , our data ( Fig 2B ) reveal a weak , albeit significant , reduction in scores signifying only a small decrease in cellular infiltration ., The reason for this apparent dissociation in the two functional properties of TNF-α in this model is unclear , and may relate to our use of a human TNF-α blocker in pigs or possibly a differential effect of TNF-α concentration on the two processes ., Interestingly , the effect of TNF-α blockade on parasite damage , as reflected in the cyst wall damage scores ( Fig 2C ) , suggests that the inhibition of measured proinflammatory , regulatory and other molecules did not inhibit damage to cysts caused by PZQ , as was found with DEX pretreatment ., ETN , a licensed biologic , has been used for TNF-α blockade for over 20 years 64 , 65 and has shown remarkable efficacy as an anti-inflammatory agent in rheumatoid arthritis and inflammatory bowel diseases 62 , 66; its safety profile is well known ., Our data demonstrate that TNF-α blockade induces potent suppression of post-treatment pericystic inflammation in a natural infection model of NCC ., The inhibitory effect of TNF-α in this model was comparable to that of DEX , a potent inhibitor of inflammation in many settings ., This study provides proof of principle that TNF-α blockade , used alone or as a steroid-sparing agent , may be a viable strategy for management of post-PZQ pericystic inflammation .

Introduction, Methods, Results, Discussion

Neurocysticercosis ( NCC ) is an infection of the brain with the larval cyst of the tapeworm , Taenia solium ., Cysticidal treatment induces parasite killing resulting in a post inflammatory response and seizures , which generally requires corticosteroid treatment to control inflammation ., The nature of this response and how to best control it is unclear ., We investigated the anti-inflammatory effects of pretreatment with etanercept ( ETN ) , an anti-tumor necrosis factor agent , or dexamethasone ( DEX ) , a high potency corticosteroid , on the post treatment inflammatory response in naturally infected pigs with neurocysticercosis after a single dose of the cysticidal drug praziquantel ( PZQ ) ., We followed the methods from a previously developed treatment model of NCC in naturally infected swine ., The four study groups of infected pigs included 3 groups treated with PZQ on day 0: PZQ-treated alone ( 100 mg/kg PO; n = 9 ) , pretreated with dexamethasone ( DEX , 0 . 2 mg/kg IM administered on days -1 , +1 and +3; n = 6 ) , and pretreated with etanercept ( ETN , 25 mg IM per animal on days -7 and 0; n = 6 ) ., The fourth group remained untreated ( n = 3 ) ., As measured by quantitative RT-PCR , ETN pretreatment depressed transcription of a wide range of proinflammatory , regulatory and matrix protease encoding genes at 120 hr post PZQ treatment in capsules of cysts that demonstrated extravasated Evans Blue ( EB ) ( a measure of blood brain barrier dysfunction ) compared to animals not receiving ETN ., Transcription was significantly depressed for the proinflammatory genes tumor necrosis factor ( TNF ) -α , and interferon ( IFN ) -γ; the inflammation regulating genes cytotoxic T-lymphocyte-associated protein ( CTLA ) 4 , interleukin ( IL ) -13 and transforming growth factor ( TGF ) -β; the tissue remodeling genes matrix metalloprotease ( MMP ) 1 and 9 , tissue inhibitors of metalloproteases ( TIMP ) 1 and 2 , and the genes regulating endothelial function vascular endothelial growth factor ( VEGF ) 1 , angiopoietin ( Ang ) 1 , Ang 2 , and platelet endothelial cell adhesion molecule ( PECAM ) -1 ., In contrast , transcription was only modestly decreased in the DEX pretreated pigs compared to PZQ alone , and only for TNF-α , IL-6 , IFN-γ , TGF-β and Ang1 ., IL-10 was not affected by either ETN or DEX pretreatments ., The degree of inflammation , assessed by semi-quantitative inflammatory scores , was modestly decreased in both ETN and DEX pretreated animals compared to PZQ treated pigs whereas cyst damage scores were moderately decreased only in cysts from DEX pretreated pigs ., However , the proportion of cysts with EB extravasation was not significantly changed in ETN and DEX pretreated groups ., Overall , TNF-α blockade using ETN treatment modulated expression of a large variety of genes that play a role in induction and control of inflammation and structural changes ., In contrast the number of inflammatory cells was only moderately decreased suggesting weaker effects on cell migration into the inflammatory capsules surrounding cysts than on release of modulatory molecules ., Taken together , these data suggest that TNF-α blockade may provide a viable strategy to manage post-treatment pericystic inflammation that follows antiparasitic therapy for neurocysticercosis .

Infection of the brain with larvae of the tapeworm Taenia solium is called neurocysticercosis ( NCC ) , a disease with varied and serious neurological symptoms ., Therapy requires antiparasitic drugs and corticosteroids to prevent seizures caused by treatment due to inflammation around dying parasites ., The gene expression of the proinflammatory molecule tumor necrosis factor alpha ( TNF-α ) is increased in NCC ., We treated three groups of naturally infected pigs with an antiparasitic drug: one group was also pretreated with an anti-TNF-α inhibitor , the second one with a corticosteroid , and the third was not pretreated ., All pigs were infused with Evans blue dye ( EB ) , which leaks where the blood brain barrier is damaged by inflammation around cysts ., We compared the expression of several genes involved in inflammation , healing and fibrosis and regulation of vascular function in tissues surrounding cysts ., In inflamed samples showing leaked EB , the inhibition of TNF-α suppressed nearly all the genes assessed , and this suppression was significantly stronger than the moderate decrease caused by corticosteroid pretreatment on most of the genes ., On microscopic examination , the inflammation observed was slightly decreased with both pretreatments in relation to the group that was not pretreated ., We believe that the inflammatory route that includes TNF-α should be further explored in the search for better management of inflammation directed to degenerating cysts .

inflammatory diseases, innate immune system, medicine and health sciences, immune physiology, cytokines, pathology and laboratory medicine, pig models, gene regulation, immunology, vertebrates, animals, mammals, animal models, developmental biology, signs and symptoms, pharmaceutics, experimental organism systems, molecular development, research and analysis methods, swine, inflammation, gene expression, immune response, immune system, eukaryota, diagnostic medicine, physiology, genetics, biology and life sciences, drug therapy, amniotes, organisms

journal.pgen.1004957

A Nitric Oxide Regulated Small RNA Controls Expression of Genes Involved in Redox Homeostasis in Bacillus subtilis

Small regulatory RNAs ( sRNA ) have been shown to play key roles in the regulation of a wide variety of cellular processes in bacteria , including stress responses , environmental signaling and virulence 1 , 2 ., They generally regulate at the post-transcriptional level by altering mRNA translation or stability ., Most sRNAs identified to date base pair with the 5’ untranslated region ( 5’-UTR ) and alter ribosome binding to the mRNA ., Changes in translation rates often have indirect consequences for mRNA stability as ribosomes can shield mRNA from attack by ribonucleases ., A number of sRNAs have also been shown to directly affect mRNA stability without altering translation initiation rates through interactions with the 5’-UTR , the 3’-UTR or the coding sequence 3 , 4 , 5 , 6 ., Although bacterial sRNAs have been studied most extensively in Escherichia coli and closely related organisms , the link to virulence has led to the identification and characterization of sRNAs in a wide range of both Gram-negative and Gram-positive bacterial pathogens ., The Gram-positive model organism Bacillus subtilis trails conspicuously behind in these efforts , where only two trans-acting sRNAs , SR1 and FsrA , have been studied in detail 7–11 ., The RNA chaperone Hfq has been shown to play a key role in sRNA association with its mRNA target in Proteobacteria ., However , its role in Firmicutes seems to be less evident 7 , 8 , 12–14 , suggesting that alternative RNA chaperones remain to be discovered in these organisms ., Furthermore , there are important differences in the mRNA degradation machineries and pathways of these two bacterial clades , most notably the widespread occurrence of a 5’-3’ exoribonuclease activity provided by RNase J in the Firmicutes and the ability of stalled ribosomes to protect long stretches of downstream mRNA from ribonucleolytic attack 15 , 16 ., The RNases involved in the regulation of mRNA stability by sRNAs in the Firmicutes have not been identified in many cases ., RsaE was first discovered in Staphylococcus aureus as a member of a family of sRNAs that contain multiple C-rich regions ( CRR ) that can potentially pair with the G-rich Shine Dalgarno ( SD ) sequences of ribosome binding sites to inhibit translation 17 ., RsaE shows some strain-dependence in its expression patterns 17 , 18 , but in all tested clinical isolates expression of RsaE was maximal during mid-exponential growth and declined in late-exponential/pre-stationary phase 19 ., Expression of RsaE in S . aureus strain RN6390 was activated by the agr quorum sensing system that plays a key role in S . aureus virulence 17 and was further shown to be induced by both oxidative stress and high salt conditions 17 , 18 ., Transcriptome and proteome analysis of RsaE deletion strains or overexpressing strains pointed to a role for S . aureus RsaE in governing the expression of genes involved in central metabolism , notably folate metabolism and the TCA cycle 17 , 18 ., RsaE is highly conserved between Bacillus and Staphylococcus species at both the primary sequence and predicted secondary structure level 17 ( Fig . 1A ) ., The two best-studied representatives of these groups , B . subtilis and S . aureus , occupy very different ecological niches , the soil and the mammalian skin and respiratory tract , respectively ., In these environments , both organisms frequently encounter nitric oxide ( NO ) , a key signaling molecule in both bacteria and eukaryotes ( for review , see 20 ) ., Indeed NO , which is toxic at high doses through the production of reactive nitrogen species ( RNS ) , is produced primarily by denitrifying bacteria in the soil and by macrophages in the mammalian host , but some species , notably Bacilli , Staphylococci and Streptomyces , can also synthesize NO via bacterial NO synthases ( bNOS ) 21 ., NO has been shown in a number of bacteria to provide protection from oxidative stress , provoked either by peroxide 22 , 23 , 24 or antibiotics 25 , 26 ., The beneficial effects of NO can also be shared between bacteria and their hosts; NO produced by B . subtilis in the intestine of C . elegans has been shown to increase the lifespan of the nematode 27 ., Despite its importance as both a signaling and potentially stress-inducing molecule , no bacterial sRNA that responds to NO levels has been identified to date ., Given that B . subtilis is a non-pathogenic organism that occupies a very different niche to S . aureus , we were curious as to the physiological role and the targets of this sRNA in B . subtilis ., We found that expression of RsaE , which we have renamed RoxS in B . subtilis for related to oxidative stress , is induced by NO in both B . subtilis and S . aureus ., Despite their similarity in sequence and regulation in the two organisms , the genes affected by deletion of this sRNA are mostly different ., Our data illustrate how the functions of a highly conserved sRNA have evolved in distantly related bacteria ., The chromosomal context of the S . aureus rsaE and B . subtilis roxS genes is very similar and , interestingly , many of the genes have functions related to redox homeostasis or show increased expression under conditions of diamide or peroxide-induced oxidative stress in B . subtilis ( S1 Fig . ) 28 ., An alignment of the homologous roxS/rsaE genes from several Bacilli and Staphylococci showed significant sequence conservation in the promoter region ( S2 Fig . ) ., An examination of a conserved 8-nucleotide ( nt ) sequence around position −65 suggested that ResD , the response regulator of the two-component system ( TCS ) ResDE , that is sensitive to both O2 and NO levels 29 , 30 , might recognize this promoter region ., Indeed , the sequence upstream of the roxS promoter is highly similar to the validated ResD binding site found upstream of the yclJ gene 31 ., We therefore tested whether the expression of RoxS was altered in a mutant lacking the ResDE TCS ., In mid-log phase , a ResDE deletion strain showed a three to four-fold decrease in RoxS expression and this effect was complemented by a plasmid expressing ResDE under IPTG control ( Fig . 1B ) , indicating that ResD is an activator of RoxS transcription ., The effect of the ResDE deletion was even stronger as the cells progressed towards stationary phase , confirming its importance as a regulator of RoxS expression ( Fig . 1C ) ., In agreement with our data , a recent chromatin immunoprecipitation study has shown a ResD binding at this location of the B . subtilis chromosome 32 ., In contrast , the thiol specific oxidative stress regulator Spx , also shown to bind in this region 33 , had little effect on RoxS expression under the conditions tested ., The membrane-bound ResE sensor kinase responds to either decreased dissolved O2 or increased NO levels 34 by a mechanism that is still not completely understood ., It then activates the ResD response regulator through phosphorylation ., We tested the effect of NO on RoxS expression by adding spermine NONOate to growing cultures ., Spermine NONOate dissolves at neutral pH with a half-life of about 39 mins to produce NO ., Expression of RoxS decreased slightly before increasing to a peak 30 mins to 1 h after addition of spermine NONOate ( Fig . 2A ) ., Although RoxS expression also decreased slightly upon addition of spermine NONOate to the ΔresDE mutant strain , no significant increase was observed after 1 h of incubation ., Thus , the NO-dependent induction of RoxS expression depends on the ResDE TCS ., Given the strong conservation of the predicted ResD binding site upstream of RsaE in Staphylococci , we asked whether expression of RsaE was subjected to a similar regulation in S . aureus ., The S . aureus homolog of the ResDE is called SrrAB and this TCS is also known to respond to low O2 levels and NO 30 ., The effect of NO on RsaE expression was also tested by adding spermine NONOate under identical conditions to those described for B . subtilis ., As for B . subtilis RoxS , we observed a weak but significant increase in RsaE expression about 1 h after the addition of spermine-NONOate to the growth medium in the wild-type ( WT ) strain ( Fig . 2B ) ., As anticipated , expression of RsaE was significantly lower in steady state conditions in the srrAB mutant , suggesting that SrrA is an activator of RsaE transcription ., Furthermore , expression of RsaE was no longer induced by the addition of spermine-NONOate in this strain , clearly showing that the induction of RsaE by NO is dependent on the SrrAB TCS ., Therefore , the signaling pathway and the expression of S . aureus RsaE and B . subtilis RoxS have been maintained during evolution ., To get insight into the regulatory role ( s ) of RoxS in B . subtilis , we performed both proteome and transcriptome analysis in strains lacking RoxS ., We detected 1092 proteins in whole cell extracts by LC-MS/MS and identified 63 proteins with significantly increased levels in the ΔroxS strain compared to the WT strain ( ≥1 . 5-fold increase by two independent methods of analysis: Spectral Counting and MS1 Filtering; S1 Table ) ., No proteins showed significantly decreased expression in the ΔroxS strain relative to WT ., Nineteen of the up-regulated candidates ( 30% ) had functions related to oxidation-reduction processes ( Fig . 3 ) , a significant enrichment from the 8 . 5% of B . subtilis genes associated with this Gene Ontology ( GO ) term ( GO:0055114 ) genome-wide ., A number of the candidates showing increased expression in the ΔroxS strain ( and therefore down-regulated by RoxS ) are predicted to be involved in oxidative stress protection ., These include the putative peroxiredoxins Tpx , AhpC and YgaF , and the thioredoxin-like protein YdbP ., We also observed increased expression of the peptide methionine sulfoxide reductase MsrA , the DNA-protecting ferritin Dps and two heme-degrading mono-oxygenases HmoA and HmoB , all of which have been shown involved in resistance to oxidative stress in different Bacilli 35–37 ., The increased expression of these proteins suggests that cells lacking RoxS are either experiencing , or behave as if they are experiencing oxidative stress , in conditions that are not normally stressful for WT cells ., Eleven of the proteins showing increased expression in the ΔroxS strain use prosthetic groups ( NAD , FAD , FMN , heme , iron-sulfur clusters ) for their oxido-reduction/electron transfer reactions ( Fig . 3 and S1 Table ) ., These include the short-chain flavodoxins YkuN and YkuP ., YkuN has been shown to be capable of transporting electrons to B . subtilis nitric oxide synthase ( bsNOS ) to generate NO from arginine 38 ., Interestingly , five of the proteins showing increased levels in the RoxS deletion mutant were members of the ferric uptake regulator ( Fur ) regulon , YkuN , YkuP , HmoA , YcgT and FeuA ( iron hydroxamate binding lipoprotein ) , and four were members of the general stress sigma B ( SigB ) regulon , YdbP , Dps , YtkL ( a predicted metal hydrolase ) and SigB itself ., One of the proteins that showed the greatest increase in expression levels was PpnKB , an inorganic polyphosphate/ATP-NAD kinase that converts NAD+ to NADP+ ., Although this enzyme is not directly involved in a redox reaction , it does have an influence on the cell’s levels of reducing power through the production of NADPH ., The ppnkB mRNA is predicted to be a direct target for RoxS repression ( see below ) ., The transcriptome analysis was performed using tiling arrays with 22 nt resolution as described previously 39 ., A comparison of the RoxS deletion strain to the WT parental strain showed 46 mRNAs with increased expression levels and 48 with decreased synthesis ( ≥2-fold; q-value <0 . 05; S2 Table ) ., Most ( 28/48 ) of the genes with decreased expression levels in the deletion strain were from the PBSX prophage , including all 12 members of the sigma factor Xpf regulon ., Nine of the genes with augmented expression in the ΔroxS strain were members of the Fur regulon , consistent with the proteome data , although some of the genes concerned were different ( Fig . 3 ) ., They include the yxeB and yusV genes , involved in the acquisition of iron , the dhbABCE operon involved in siderophore biosynthesis and the flavodoxin-encoding ykuNOP operon ., YkuN and YkuP were among six candidates also identified in the proteome analysis , the others being Tpx , LplJ ( lipoate protein ligase ) , YhfE ( putative endogluconase ) and YktB ( unknown ) ., We confirmed the increase in ykuNOP expression in the ΔroxS strain by Northern blot ( S3A Fig . ) although we suspect it may be an indirect consequence of the roxS deletion on Fur activity ( see below ) ., Furthermore , genes encoding a thioredoxin ( resA ) and the putative peroxiredoxin ( tpx ) also showed increased expression in the absence of RoxS ., The resA gene encodes an extracytoplasmic thioredoxin involved in the maturation of cytochrome C , while the function of Tpx is still unknown ., Two genes involved in the pentose phosphate pathway , tkt , encoding transketolase and gndA , encoding 6-phospho-gluconate dehydrogenase , also showed increased transcript levels ., The pentose phosphate pathway is a major source of NADPH production in the cell for use as a reducing agent in anabolic reactions such as lipid and nucleic acid synthesis ., Overall , fourteen of the up-regulated genes in the tiling array ( 30% ) had annotated functions related to oxidation-reduction reactions ( GO term 0055114 ) , consistent with the functional enrichment seen in the proteome study ( Fig . 3 and S2 Table ) ., These data are in good agreement with a general role for RoxS in the redox state/oxidative stress response ., Because most sRNAs base-pair to mRNA targets , we used several programs ( TargetRNA2 40 , CopraRNA 41 , RNApredator 42 ) to predict potential direct mRNA targets of RoxS , with a particular focus on the translation initiation region ., The targets suggested by CopraRNA were highly enriched for mRNAs involved in a relatively small number of cellular processes including electron transport , respiration , lipid metabolism and metal binding ( S4 Fig . ) ., The best target proposed by TargetRNA2 and RNA predator , the ppnKB mRNA ( Fig . 4A ) , was consistent with this functional enrichment ., Furthermore , the synthesis of the PpnKB protein was 4 . 5 to 9-fold increased in the ΔroxS strain by Spectral Counting and MS1 Filtering , respectively ( Fig . 3 ) ., We therefore chose to study the RoxS-dependent regulation of ppnkB in more detail ., To determine whether RoxS could directly bind to the ppnKB mRNA to inhibit translation initiation , we tested the effect of RoxS on the formation of the ribosomal initiation complex on the ppnKB mRNA by toeprinting assays ., Addition of 30S ribosomal subunits and initiator tRNA to the ppnKB transcript , showed a clear toeprint at position +16 relative to the ppnKB start codon ( Fig . 4B; lane 5 ) ., Incubation of the ppnKB transcript with equimolar and higher concentrations of RoxS resulted in complete inhibition of the 30S ribosome toeprint , while a band specific to the binding of RoxS appeared at position −9/10 ( Fig . 4B; lane 6 and 7 ) ., In contrast , RoxS had a much weaker effect on the formation of the initiation complex on the ykuN transcript ( S3B Fig . ) and did not show evidence for a stable interaction around the SD sequence , consistent with the fact that , despite the presence of four consecutive G-residues in the SD , it was not predicted as a target by any of the three algorithms ( including the ORFs , for TargetRNA2 ) ., This experiment shows that RoxS specifically binds to the ppnKB mRNA and forms a stable complex that is sufficient to prevent the formation of the ternary translation initiation complex ., The toeprinting assays , coupled with the fact that RoxS is not predicted to make significant interactions with any portion of the ykuNOP mRNA , suggest that RoxS-dependent effect on the expression this operon , observed in both the transcriptome and proteome analysis , most likely results from an indirect effect ., The base-pairing interaction between RoxS and ppnKB predicted by TargetRNA is extensive ( Fig . 4A ) and includes the first three C-rich regions ( CRR1-3 ) ., However , the strong reverse transcriptase ( RT ) stop at nt −9/10 provoked by duplex formation is close to the SD sequence , suggesting the most stable interaction is between CCR3 and the ppnKB ribosome binding site ., However , the six nts downstream of CRR1 are identical to those downstream of CRR3 , creating a 10 nt duplication ( CCCCUUUGUU ) in RoxS and leaving open the possibility that the two sequences were functionally redundant ., We therefore performed toeprinting assays with RoxS variants where the four consecutive C-residues of CRR1 or CRR3 , or both , were changed to A . These mutations are not predicted to alter the secondary structure of RoxS ., The data clearly showed that mutation of CRR3 alone abolished the ability of RoxS to bind the mRNA and to inhibit 30S ribosome binding to ppnKB ( Fig . 4B , lanes 14–15 ) ., Conversely , mutation of CRR1 alone had no effect ( Fig . 4B , lanes 10–11 ) ., RoxS mutants lacking both CRR1 and CRR3 , or the three CRR’s 1 , 2 and 3 , behaved similarly to the CRR3 mutant in failing to interact with the ppnKB mRNA or to inhibit translation initiation complex formation ( Fig . 4B , lanes 18–19 and 22–23 ) ., Hence , these data show that CRR3 plays the most important role in inhibition of translation initiation and that the two repeat motifs of RoxS are not functionally equivalent for the regulation of ppnKB ., Binding of sRNAs can have direct effects on target mRNA stability , by creating new sites for endoribonuclease cleavage , or indirect effects through the increased exposure of existing cleavage sites following translational repression ., We therefore asked whether overexpression of RoxS would lead to degradation of the ppnKB mRNA ., For these studies , we used the ΔroxS mutant strain transformed with a plasmid expressing RoxS from a tetracycline-dependent promoter ( strain CCB498: ΔroxS + pDG-Ptet-roxS ) or with a control plasmid ( strain CCB505: ΔroxS + pDG-Ptet ) ., Induction of RoxS expression with increasing concentrations of anhydrotetracylcine ( aTc ) in strain CCB498 caused a gradual reduction ( about two-fold ) in ppnKB mRNA levels compared to the empty vector control strain ( Fig . 5 ) , showing that RoxS affects the amount of ppnKB mRNA in the cell , in addition to controlling its translation ., Expression of RoxS from the pDG-Ptet vector is transient , reaching a peak about 5 mins after addition of aTc before decreasing rapidly ( Fig . 6A , D ) , presumably due to an accumulation of the TetR repressor driven by the same promoter ., We exploited this property of the plasmid to analyze whether ppnKB mRNA levels would recover upon shut-down of RoxS expression ., Indeed , ppnKB mRNA levels fell to a minimum about 5 mins after induction of RoxS and were rapidly restored as RoxS levels decreased ( Fig . 6A , D ) ., The RoxS-dependent reduction in ppnKB levels was only slightly less efficient in a strain lacking the double-strand specific endonuclease RNase III , encoded by the rnc gene ( Fig . 6B , D ) ., However , it was significantly reduced in a strain lacking the single-strand specific nuclease RNase Y , encoded by rny ( Figs . 6C , D ) ., These results suggest that RNase Y is a key enzyme for RoxS-mediated ppnKB mRNA turnover , while RNase III plays a secondary role under these experimental conditions ., It should be noted that RoxS is slow to shut-off in the rny mutant ( Fig . 6C , D ) ; we will see later that this is due to an effect of RNase Y on RoxS RNA stability ., To further show that RoxS controls ppnKB expression at the level of mRNA stability , we measured the half-life of the ppnKB mRNA in WT strains and mutant strains lacking either RNase III or RNase Y under steady state conditions ( Fig . 7 ) ., The ppnKB mRNA was stabilized about 1 . 8-fold in cells lacking RoxS ( 8 . 4 vs . 15 mins half life , respectively in WT and ΔroxS strains ) , consistent with a role for RoxS in controlling ppnKB mRNA stability ( Fig . 7A ) ., In the absence of RNase III , a similar increase in ppnKB stability was seen , but was not further amplified by the additional deletion of roxS ( Fig . 7B ) ., The simplest explanation is that RNase III and RoxS collaborate to degrade a portion of ppnKB transcripts; the lack of either component , or both , leading to a similar increase in ppnKB mRNA stability ., The ppnKB mRNA was also significantly stabilized ( 8 . 4 vs . 24 mins ) in the Δrny mutant compared to the WT strain ( Fig . 7C ) , consistent with a role for RNase Y initiating the degradation of the ppnKB mRNA ., In this case , however , further deletion of roxS had an additional stabilizing effect ( 24 vs . >40 mins half life , respectively ) ., This suggests the existence of a RoxS-mediated ppnKB turnover pathway that is independent of RNase Y and that inactivation of both pathways are required for maximal stabilization of ppnKB ., We propose that the second pathway is the RoxS/RNase III dependent pathway described above ., The data also indicate an effect of RNase Y that is independent of RoxS ( 15 mins vs . >40 mins half-life , respectively , in ΔroxS vs . Δrny ΔroxS strains ) , consistent with a role for RNase Y in the non-regulated turnover of the ppnKB mRNA ., Interestingly , the ppnKB mRNA was highly unstable in a strain lacking the 5′-3′ exoribonuclease RNase J1 ( Fig . 7D ) and this destabilization was attenuated upon deleting RoxS ( 3 . 1 vs . 9 . 4 mins half-life , respectively , in ΔrnjA vs . ΔrnjA ΔroxS strains ) ., Data presented in the next section will shed light on this phenomenon ., Globally , our data provide an illustration of the complex interplay between ribonucleases involved in the turnover of the ppnKB mRNA , both dependent and independent of RoxS-mediated repression ., We also analyzed the importance of the three main ribonucleases in the degradation of the RoxS sRNA ., The half-life of the chromosomal copy of RoxS was first measured in WT cells and in cells lacking either RNase III , RNase Y or RNase J1 ., In WT cells and in cells lacking RNase III , RoxS showed bi-phasic RNA degradation upon transcription arrest with rifampicin ( Fig . 8A , C ) , suggesting that two populations of this sRNA exist in vivo ., The simplest interpretation is that these populations represent free RoxS or RoxS bound to its targets ., While the half-life of the rapidly decaying population was similar in both strains , the slowly decaying population was strongly stabilized in the Δrnc mutant ( Fig . 8C ) ., Because of its specificity for double-stranded RNA , it is most likely that the stabilisation of the slowly decaying phase represents stabilisation of RoxS molecules that are hybridized to its mRNA targets ., In cells lacking RNase J1 ( ΔrnjA ) , full length RoxS was stabilized compared to the WT strain , but in addition a very long-lived degradation/processing intermediate was detected ( Fig . 8B , C ) ., This intermediate was not detected in the absence of RNase Y and the full-length RoxS had a much longer half-life in the Δrny mutant strain ( Fig . 8B , C ) ., Using primer extension , we mapped the 5’ end of the short RoxS fragment to nt +20 of RoxS ( S5 Fig . ) ., Together these results suggest that RNase Y initiates RoxS turnover by cleaving around nt +20 ( Fig . 1 ) and RNase J1 degrades the downstream cleavage product , in addition to having some activity on the full-length RNA ., Because a small amount of RNase Y-cleaved RoxS was visible in WT cells ( Fig . 8A ) , we asked whether this truncated form was functional and might contribute to regulation ., We cloned a 5′ truncated version of RoxS beginning at nt 20 , called RoxS ( Y ) , into the plasmid vector pDG-Ptet ., In a manner similar to full-length RoxS , aTc induction of RoxS ( Y ) resulted in a rapid and efficient reduction in ppnKB levels , which then recovered as RoxS ( Y ) levels fell ( S6A-S6B Fig . ) ., Thus the truncated form of RoxS that accumulates in an RNase J1 mutant is fully functional and may explain the RoxS-dependent destabilization of ppnKB in the absence of RNase J1 ( Fig . 7D ) ., When tested in the toeprinting assay , the truncated RoxS species formed a more extensive hybrid than the full-length sRNA with the ppnKB mRNA , indicated by additional reverse transcriptase stops around nt −2 and nt +23 ( Fig . 4B , lane 24 ) ., The short form was equally efficient as the full-length RoxS in inhibiting ppnKB translation initiation complex formation at the concentrations tested ., To characterize the interactions between ppnKB and RoxS and its various mutant forms , we performed structure probing experiments on the ppnKB mRNA using the double-strand-specific enzymes RNase V1 and RNase III , and RNase T1 , which cleaves principally 3′ to unpaired guanines ( Fig . 9 and S7 Fig . ) ., The data suggested that in the absence of RoxS , the ppnKB mRNA folds into a long , but relatively unstable hairpin structure that extends from nt −37 to nt +34 relative to the translation initiation site ( Fig . 9A ) ., Indeed , two major RNase T1 cleavages occur 3′ to G-3 and G-4 and two lesser cleavages 3′ to nts +2/+3 in the apical loop containing the AUG initiation codon while a number of RNase V1 cleavages are located in the irregular helix ( Fig . 9A and S7A-S7B Fig . ; lane 2 ) ., Consistent with this model , RNase III cleaves the large irregular helix of ppnKB at four sites ( nts −24 , +10 , +22 and +32 ) , with the cleavages at −24 and +22 producing the two-nt 3’ overhang characteristic of RNase III processing ( Fig . 9A and 9D; lane 3 ) ., Binding of RoxS induces strong protection of the RNase T1 cleavages in the apical loop ( G+3 , G-3 , G-4 ) and at G-12 and G-13 of the SD sequence while the cleavage at G-37 is slightly enhanced ( Fig . 9B and S7A Fig . ; lanes 3 , 4 ) ., Concomitantly , RNase V1 cleavages are slightly enhanced at nts −29/30 , +16 to +18 , +24/25 and +31 ( Fig . 9B and S7B Fig . ; lanes 3 , 4 ) ., Remarkably , all four RNase III cleavage sites are significantly reduced upon binding to RoxS while two strong adjacent cleavages appear in the ppnKB SD sequence at nts −12/13 ( Fig . 9D; lanes 4 , 5 ) ., These data suggest that the large hairpin loop of ppnkB undergoes a partial melting to promote basepairing interactions with RoxS , leading to the sequestration of the SD sequence ., Identical changes in the RNase III cleavage patterns were observed if complex formation was performed with the truncated RoxS ( Y ) ( Fig . 9C and 9D , lanes 22–25 ) or with the CRR1 mutant ( Fig . 9D; lanes 6–9 ) ., However , RoxS derivatives with a mutation in CRR3 had no effect on the RNase III cleavages , showing that the mutated RNAs fail to interact with ppnKB ( Fig . 9D ) ., Identical conclusions were reached in the probing experiments with RNases T1 and V1 ( S7 Fig . ) ., The proposed models for the interaction of ppnkB with full-length or truncated RoxS ( Fig . 9B and C ) take into account most of the data although we cannot completely distinguish between RoxS-dependent changes that are due to the formation of an extended RoxS/ppnKB duplex or due to a stabilization of existing ppnKB helices upon RoxS binding ., However , the data unambiguously show that the CCR3 motif is responsible for the interaction with the SD sequence to prevent the formation of the translation initiation complex and to create a novel site for RNase III binding and cleavage ., TargetRNA2 and CopraRNA both predicted the sucC gene , the first cistron of the sucCD operon encoding the two subunits of succinyl-coA synthase , as another potential target of RoxS ( Fig . 10A ) ., Differential proteomic analysis showed a 2-fold increased expression of SucD in the mutant ΔroxS strain , while SucC narrowly missed the dual 1 . 5-fold cut-off ( 1 . 4 fold increase by spectral counting; 1 . 8 fold increase by MS filtering ) ( Fig . 3 , S1 Table ) ., Interestingly , the sucCD mRNA was also shown to be a target of RsaE in S . aureus 17 , 18 ., We therefore probed the membranes shown in Fig . 6 for the sucCD mRNA to see whether its mRNA levels were affected by RoxS expression ., Transient expression of RoxS by aTc addition led to a similar decrease in sucCD expression as was observed for ppnKB ( Fig . 10B , E ) ., This decrease in expression was slightly attenuated in the absence of both RNase III ( Fig . 10C , E ) and RNase Y ( Fig . 10D , E ) , suggesting roles for both of these enzymes in the turnover of the sucCD mRNA in response to RoxS expression ., We then analyzed the effect of RoxS and its variants on the formation of the translation initiation complex formed with the sucC mRNA using toeprinting assays ., In contrast to the ppnKB mRNA , RoxS binding did not prevent the formation of the initiation complex on sucC , even at the highest RoxS concentration ( Fig . 11 , lanes 6 and 7 ) ., Only a weak RT pause characteristic of RoxS binding was observed at nt +4/5 relative to the first nt of the open reading frame , indicating that RoxS did not form a stable complex with sucC ( Fig . 11 , lane 4 ) ., To our surprise , the truncated form of RoxS ( Y ) bound far more efficiently than RoxS to the sucCD mRNA , causing a very strong RT pause at position −2/−3 and an increased signal at +4/5 ( Fig . 11 , lane 24 ) ., As a consequence , this led to a strong and efficient inhibition of the toeprint at +16 by the truncated form of RoxS ( Fig . 11 , lanes 26 and 27 ) comparable to that seen with ppnKB ( Fig . 4 , lanes 6 and 7 ) ., This experiment suggests that processing of RoxS is necessary for regulation of sucC and that truncation at the 5’ end of RoxS expands the repertoire of effective targets for this sRNA ., In this paper , we have shown that expression of the regulatory RNA RoxS/RsaE is induced by nitric oxide in both B . subtilis and S . aureus , in a mechanism that is dependent on their respective orthologous two-component systems , ResDE and SrrAB ., The membrane-bound sensor protein ResE/SrrB is autophosphorylated in response to both NO and limiting O2 levels , and in turn phosphorylates the response regulator ResD ( SrrA in S . aureus ) 31 , 43 ., NO and hypoxia inhibit terminal oxidases and limit the flow of electrons through the electron transport chain ., It was recently suggested that the resulting accumulation of reduced menaquinones in the membrane is likely to be the trigger that activates the ResDE/SrrAB TCS 30 , similar to the quinone-sensitive ArcAB TCS in E . coli 44 ., In S . aureus , SrrAB is important for cell survival in the host environment and in biofilms ., This system senses and responds to both NO and hypoxia , and regulates genes required for cytochrome biosynthesis and assembly , anaerobic metabolism , iron-cluster repair , and NO detoxification 43 ., In B . subtilis , ResD is known to activate the expression of about 30 genes involved in the anaerobic respiration of nitrate , the production of cytochromes , the fermentation of pyruvate and in NO detoxification 45 ., Here , we have studied in more detail the regulatory functions of B . subtilis RoxS , which further expands the regulatory impact of ResD ., Interestingly , regulation by ResD is significantly more efficient as growth begins to slow down and RoxS expression is essentially completely ResD-dependent in early stationary phase ( Fig . 1C ) ., Despite the fact that a number of the surrounding genes show increased expression in the presence of diamide , the thiol stress regulator Spx had little effect on RoxS expression ., This is consistent with a recent study by Rochat et al . that showed an effect of an spx deletion on the neighboring genes but not on roxS itself 33 ., However , a significant number of genes with functions related to oxidation-reduction reactions or oxidative stress resistance showed increased expression in B . subtilis cells lacking RoxS ., This surprising result suggests that ΔroxS cells are suffering from a deficit of reducing power ., The derepression of many members of the Fur regulon , including the ykuNOP o

Introduction, Results, Discussion, Materials and Methods

RsaE is the only known trans-acting small regulatory RNA ( sRNA ) besides the ubiquitous 6S RNA that is conserved between the human pathogen Staphylococcus aureus and the soil-dwelling Firmicute Bacillus subtilis ., Although a number of RsaE targets are known in S . aureus , neither the environmental signals that lead to its expression nor its physiological role are known ., Here we show that expression of the B . subtilis homolog of RsaE is regulated by the presence of nitric oxide ( NO ) in the cellular milieu ., Control of expression by NO is dependent on the ResDE two-component system in B . subtilis and we determined that the same is true in S . aureus ., Transcriptome and proteome analyses revealed that many genes with functions related to oxidative stress and oxidation-reduction reactions were up-regulated in a B . subtilis strain lacking this sRNA ., We have thus renamed it RoxS ., The prediction of RoxS-dependent mRNA targets also suggested a significant enrichment for mRNAs related to respiration and electron transfer ., Among the potential direct mRNA targets , we have validated the ppnKB mRNA , encoding an NAD+/NADH kinase , both in vivo and in vitro ., RoxS controls both translation initiation and the stability of this transcript , in the latter case via two independent pathways implicating RNase Y and RNase III ., Furthermore , RNase Y intervenes at an additional level by processing the 5′ end of the RoxS sRNA removing about 20 nucleotides ., Processing of RoxS allows it to interact more efficiently with a second target , the sucCD mRNA , encoding succinyl-CoA synthase , thus expanding the repertoire of targets recognized by this sRNA .

Bacteria have evolved various strategies to continually monitor the redox state of the internal and external environments to prevent cell damage and/or to protect them from host defense mechanisms ., These signals modify the expression of genes , allowing bacteria to adapt to altered redox environments and to maintain homeostasis ., Studies in Enterobacteriaceae have shown that sRNAs play central roles in adaptation to oxidative stress ., We show here that the conserved sRNA , RoxS is induced by the presence of nitric oxide ( NO ) in the medium , through the ResDE and SrrAB two-component systems of Bacillus subtilis and Staphylococcus aureus , respectively ., B . subtilis RoxS regulates functions related to oxidation-reduction reactions and acts as an antisense RNA to control translation initiation and the degradation of ppnKB mRNA , encoding an NAD+/NADH kinase ., Interestingly , RNase Y processes the 5′ end of the RoxS sRNA leading to a truncated sRNA that in turn interacts more efficiently with a second target , the sucCD mRNA , encoding succinyl-CoA synthase ., Taken together this work shows that RoxS is part of a complex regulatory network that allows the cell to sense and respond to redox perturbations , and revealed a novel process that allows an expansion of the repertoire of sRNA targets .

journal.ppat.0030185

Bradykinin B2 Receptors of Dendritic Cells, Acting as Sensors of Kinins Proteolytically Released by Trypanosoma cruzi, Are Critical for the Development of Protective Type-1 Responses

Chagas disease , the chronic cardiomyopathy caused by infection with the intracellular parasitic protozoan Trypanosoma cruzi , remains a major health problem in Central and South America 1 ., Although acute Chagas disease may have a fatal outcome , the blood parasitemia , tissue parasite burden ( liver , spleen , and heart ) , and the inflammatory sequel tend to subside with the onset of adaptive immunity ., After several years of asymptomatic infection , approximately 30% of infected patients develop a chronic and progressive form of cardiomyopathy 2 ., While not excluding a secondary pathogenic role for autoimmunity , studies in humans and animal models support the concept that parasite persistence in myocardial tissues is the primary cause of chronic immunopathology 3–6 ., Cohort studies with chagasic patients have linked chronic heart pathology to TH1-type responses 7 , but this proposition was recently called into question by a report indicating that the frequency of IFN-γ-producing effector/memory T cells is inversely correlated with the severity of chronic Chagas disease 8 ., Animal model studies established that acquired resistance depends on development of serum antibodies as well as on IFN-γ-producing CD4+ and CD8+ T cells 9–12 ., Recent studies indicated that CCR5 has a suceptible phenotype , attributed to impaired recruitment of effector T cells to parasitized heart tissues 13 , 14 ., Although the dominant epitope specificities recognized by cytotoxic CD8 T cells are encoded by highly polymorphic genes 15 , it is still unclear how T . cruzi escapes from immune surveillance 16–18 ., In the present work , we set out to investigate the mechanims linking innate to adaptive immunity in the mouse model of T . cruzi infection ., Early studies about innate resistance mechanisms indicated that macrophages upregulate nitric oxide ( NO ) -dependent trypanocydal responses 19 due to ligand-induced signaling of Toll-like 2 receptors ( TLR2 ) 20 , 21 or TLR4 22 ., More recently , Bafica et al . reported that macrophages sense T . cruzi DNA via triggering of intracellular TLR9 23 ., Interestingly , they showed that acute infection is more severe in TLR2−/− TLR9−/− mice than in TLR9−/− mice or either TLR2−/−- 23 or TLR4-deficient mice 22 , albeit not as much as in the overtly susceptible MyD88−/− mice 24 ., While not formally excluding an additive innate role for TLR4 , these collective studies suggested that cooperative activation of TLR2 and TR9 may account for the bulk of protective IFN-γ responses generated by MyD88-dependent signaling pathways 23 , 24 ., Of note , analysis of macrophage activation by MyD88-independent pathways revealed that TLR/TRIF coupling promotes NO-dependent microbicidal responses through upregulation of type I interferons 25 , 26 ., In spite of evidence that mice deficient in IL-12 27 are highly susceptible to T . cruzi infection , it is still uncertain if induction of TH1-responses is strictly dependent on dendritic cell ( DC ) maturation by TLRs/MyD88-dependent pathways ., Pertinently , it was reported that spleen cells from MyD88−/− mice display small yet significant production of IL-12 and IFN-γ 24 , 28 ., These observations imply that IL-12-dependent Th1 responses may be also controlled by MyD88-independent mechanisms , such as the NKT/CD1d pathway 29 , or by endogenously released bradykinin ( BK ) , an endogenous danger signal driving DC maturation 30–32 ., “Kinins” , a small group of mediators related to the nonapeptide BK , activate immature DCs 30 as well as several other cell types through the binding to distinct subtypes of G-protein-coupled receptors: B2R ( constitutive ) and B1R ( inducible ) 33–36 ., The B2R agonists , BK or lysyl-BK ( LBK ) , are proteolytically excised from an internal segment of their parental ( glyco ) proteins , high or low molecular weight kininogens , by plasma or tissue kallikreins , respectively 33 ., In the settings of infections , however , kinins can be generated through the direct action of microbial cysteine proteases , such as gingipain of Porphyromonas gingivalis 37 and cruzipain ( CZP ) , the major cysteine protease of T . cruzi 38–41 ., Using a subcutaneous model of T . cruzi infection , we recently demonstrated that trypomastigotes release kinins in peripheral tissues through the activity of CZP 31 ., Once liberated from plasma borne–kininogens , the short-lived kinin peptides activate CD11c+DCs via B2R , inducing IL-12 production and stimulating the migration of these antigen-presenting cells ( APCs ) from the periphery to the draining lymph nodes , where they initiate TH1-like responses against T . cruzi 31 , 32 ., Here we report that B2R-deficient mice infected intraperitoneally by T . cruzi display a typical susceptible phenotype ., Adoptive cell transfer experiments demonstrate that CD11c+ DCs activated by the endogenous kinin/B2R-signaling pathway are critically required for the induction and/or maintenance of activated effector CD4+ and CD8+ T cells , while limiting the development of potentially detrimental IL-17-producing CD4+ T cell ( TH17 ) responses in mice acutely infected with T . cruzi ., In order to test the hypothesis that kinins may contribute to immune control of T . cruzi infection 30 , 31 , we injected intraperitoneally B2R+/+ C57BL/6 and B2R−/− mice with tissue culture trypomastigotes ( TCT ) of either Dm28c strain ( 1 × 106 ) or Brazil strain ( 1 × 104 ) ., The data shown in Figure 1 indicate that wild-type mice infected with Dm28c TCT developed a low blood parasitemia and all the animals survived ( Figure 1A , higher panel ) ., In contrast , B2R−/− mice infected with Dm28c showed a precocious blood parasitemia ( day 13 post-infection p . i . ) , which further increased ( approximately 3-fold ) as the infection continued ( 23 d p . i . ) ., Mortality rates indicated that B2R−/− mice infected by Dm28c TCT started to die earlier ( day 16 ) than wild-type mice and were all dead by day 27 ( Figure 1A , lower panel ) ., We then studied the outcome of infection with the Brazil strain ., The results ( Figure S1 ) show that wild-type mice displayed a relatively low blood parasitemia and the mortality rate did not exceed 20% ., In contrast , the B2R−/− mice infected by Brazil strain developed increased blood parasitemia , and 80% of these animals were dead by day 28 ( Figure S1 ) ., We then further characterized the outcome of intraperitoneal infection with the Dm28c strain , using a lower inoculum ( 6 × 105 ) ., Analysis by real-time PCR ( qPCR ) showed that heart tissues of infected B2R−/− mice ( 14 d p . i . ) contained approximately 5-fold higher content of parasite DNA as compared to wild-type heart ( Figure 1B ) ., Surprisingly , we found that the parasite tissue burden in the spleen was very low both in B2R+/+ ( 0 . 30 ± 0 . 09 fg/100 ng host DNA ) and B2R−/− ( 0 . 46 ± 0 . 21 fg/100 ng host DNA ) mice ( Figure 1B ) ., Thus , unlike the scenario observed in extra-lymphoid tissues , parasite outgrowth in the spleen is controlled by mechanisms that do not critically depend on activation of the kinin/B2R pathway , at least so at relatively early stages ( 14 d ) of infection ., Since the tissue parasitism in the spleen of wild-type and B2R−/− mice ( 14 d p . i . ) was marginal , we checked whether type-1 effector cells were generated in lymphoid tissues of both mice strains ., Recall assays indicated that splenocytes from wild-type or B2R−/− vigorously secreted IFN-γ upon stimulation with soluble T . cruzi antigen ( Ag ) ( Figure 2A ) ., Controls showed that , in the absence of T . cruzi soluble Ag , there was no significant production of IFN-γ by the splenocytes ( Figure 2A ) ., We then scrutinized the ex vivo recall responses of CD4+ or CD8+ T cells derived from either wild-type or B2R−/− spleen ( isolated from infected or naïve mice , as controls ) using wild-type CD11c+ DCs ( purified from normal spleen ) as APCs , to exclude the possibility that eventual defects in Ag processing/presentation by B2R−/− DCs could interfere with our “read-outs” ., In keeping with the potent type-1 response elicited by unfractionated wild-type and B2R−/− splenocytes ( 14 d p . i . ) , fluorescent activated cell sorting ( FACS ) analysis showed presence of high and comparable frequencies ( Figure 2B , lower panel ) of IFN-γ-producing CD4+ and CD8+ T cells in the spleens of wild-type and B2R−/− mice ( Figure 2B ) ., Controls performed with Ag-stimulated CD4+ or CD8+ T cells isolated from naïve mice did not generate significant frequencies of IFN-γ-producing cells ., Consistent with the similar FACS profiles , ELISA assays showed that IFN-γ was vigorously secreted by Ag-responsive splenic CD4+ or CD8+ T cells , irrespective of the mouse strain origin ( Figure 2C ) ., We then checked if the presence of type-1 CD4+ and CD8+ effector T cells was maintained in the spleen as the infection continued ., Recall assays performed 2 wk later ( 28 d p . i . ) indicated that IFN-γ production by wild-type splenocytes remained vigorous , while the type-1 response of Ag-stimulated B2R−/− splenocytes declined sharply ( Figure 3A ) ., We then repeated this analysis using CD4+ or CD8+ T cells purified from the spleens of infected wild-type mice or B2R−/− mice , using wild-type DCs as APCs ., Consistent with the data obtained with splenocytes , we found that Ag-stimulated T lymphocytes ( CD4+ or CD8+ ) isolated from B2R−/− spleen ( 28 d p . i . ) secreted significantly lower levels of IFN-γ as compared to wild-type splenic T cells ( unpublished data ) ., We then performed FACS analysis to further characterize the phenotypic changes that occurred in the spleen , as the acute infection advanced ( 28 d p . i . ) ., Our results ( Figure 3B ) showed that Ag-stimulated T cells isolated from wild-type spleen showed high frequencies of IFN-γ-producing CD4+ and CD8+ T lymphocytes ., Moreover , a significant fraction of activated CD4+ and CD8+ T cells isolated from spleen of wild-type infected mice displayed the CD44 surface marker ., As expected , addition of Ag to CD4+ or CD8+ T cell cultures from naïve mice did not lead to IFN-γ production ( Figure 3B , lower panel ) ., In contrast , B2R−/− spleen presented low frequencies of IFN-γ-producing CD4+ or CD8+ effectors ( CD44− ) ( Figure 3B ) ., Although we have no direct evidence that the Ag-responsive T cells detected ex vivo include functionally active effectors , it is worthwhile mentioning that adoptive transfer of CD4+/CD8+ T cells ( isolated from wild-type mice at 60 d p . i . ) into B2R−/− mice rendered these recipient mice resistant to lethal infection ( 0% mortality , n = 5; three independent experiments ) , as compared to non-manipulated B2R−/− mice ( 100% mortality ) or B2R−/− mice that received CD4+/CD8+ T cells from normal wild-type mice ( 100% mortality ) ., As mentioned earlier in this section , we found a 5-fold increase of T . cruzi DNA in the heart of B2R-deficient mice at day 14 p . i . , as compared to wild-type heart ( Figure 1C ) ., In view of these findings , we set out to determine if cardiac tissues of wild-type and B2R−/− mice contained type-1 effector T cells ., Recall assays ( again using wild-type splenic CD11c+ DCs as APCs ) showed that IFN-γ production by intracardiac B2R−/− CD4+ T cells was significantly diminished ( over 50% ) as compared to responses elicited by experienced CD4+ T lymphocytes isolated from wild-type heart at 14 d p . i . ( p < 0 . 01 ) ( Figure 4 ) ., Similarly , the initial recall response of intracardiac CD8+ T cells isolated from B2R−/− mice was approximately 60% lower than that of wild-type CD8+ T cells ( Figure 4 ) ., We then checked if the type-1 cytokine response of intracardiac T cells from B2R−/− mice was further compromised as the infection continued ., The FACS profiles of wild-type-infected mice ( 28 d p . i . ) revealed high frequencies of IFN-γ-producing intracardiac CD4+ and CD8+ T cells ( Figure 5 ) ., In addition , we found that the CD44 marker characteristic of activated T cells was present in a significant proportion of wild-type intracardiac CD4+ T cells , and ( to lower extent ) also in the CD8+ T cell subset ( Figure 5 , upper and lower panels ) ., In contrast , B2R−/− mice exhibited very low frequencies of CD4+ and CD8+ T cells in the intracardiac CD3+ T cell pool at day 28 p . i . ( Figure 5 ) ., Following the same trend , IFN-γ-producing CD4+ or CD8+ effector T cells , and activated phenotypes ( CD44+CD4+ and CD44+CD8+ T cells ) were virtually absent from B2R−/− heart ., Collectively , these results suggest that activation of the endogenous kinin/B2R signaling pathway in T . cruzi–infected mice may have an impact on the control mechanisms affecting the temporal and spatial activity of type-1 effectors ., Considering that the type-1 responses of B2R−/− mice were depressed both in the heart ( as early as 14 d p . i . ) and spleen ( 28 d p . i . ) , we then asked if these effects were coupled to TH2 upregulation ., Our results indicated that Ag-stimulated T CD4+ T cells ( isolated from B2R−/− heart or spleen ) did not upregulate IL-4 production ( unpublished data ) ., Since IFN-γ inhibits TH17 lineage development in vitro 42 , 43 , we wondered if the reduced TH1 responses observed in B2R−/− mice were accompanied by rises of IL-17- and TNF-α-producing T cells ., Recall responses made at 28 d p . i . ( Figure 6A ) revealed that splenic CD4+ T lymphocytes from wild-type mice did not secrete significant levels of IL-17 , while splenic B2R−/− CD4+ T cells upregulated IL-17 ., The same trend was found when we measured TNF-α levels secreted by experienced B2R−/− CD4+ T cells ( Figure 6B ) ., Similar data were obtained when we compared Ag-stimulated responses of intracardiac CD4+ T cells isolated from B2R−/− versus wild-type mice , as discussed later in this section ., Collectively , these data suggest that the TH17/TH1 ratio was drastically increased as the acute infection advanced in the highly susceptible B2R−/− mice ., Since type-1 responses were impaired in infected B2R−/− mice , we sought to determine if IL-12 responses were preserved , or not , in these mutant mice ., To this end , we inoculated Dm28c TCT ( 1 × 106 ) intravenously in wild-type and B2R−/− mice , isolated splenic CD11c+ DCs 18 h p . i . , and measured IL-12 production by FACS ., The results ( Figure 7A ) showed a marked increase in the frequency of IL-12-producing CD11c+ DCs ( 8% ) in B2R+/+ in relation to non-infected controls ( no IL-12 staining ) ., In contrast , splenic CD11c+ DCs isolated from infected B2R−/− mice showed a low frequency ( 2% ) of IL-12-positive cells ( Figure 7A ) ., These results were corroborated by ELISA determinations of IL-12 responses produced by DCs isolated from intravenously infected mice ( Figure 7B ) ., Of note , we found that macrophages ( CD11b+ F4/80+ ) from infected wild-type and B2R−/− mice show enhanced production of IL-12 as compared to naïve mice , suggesting that alternative mechanisms ( i . e . , B2R-independent ) may govern IL-12 production by splenic macrophages ( unpublished data ) ., Extending these in vivo studies to BALB/c mice , these animals were pre-treated , or not , with the B2R antagonist HOE-140 before intravenous injection of TCT ., The FACS profiles showed a sharp increase of IL-12-positive CD11c+ DCs in BALB/c mice injected with either TCT ( Figure S2 ) or BK ( positive control ) ( Figure S2 ) ., In contrast , BALB/c mice pre-treated with HOE-140 showed a reduced frequency of IL-12-positive CD11c+ DCs ( Figure S2 ) ., Collectively , the data indicate that B2R drives IL-12 production by splenic DCs , at least at very early stages of the infection ., We then carried out in vitro studies to verify if the parasites could induce the maturation of CD11c+ ( splenic ) DCs through the activation of the kinin/B2R signaling pathway ., IL-12 production and surface expression of co-stimulatory proteins were used as read-out for DC maturation ., FACS analyses showed that CD11c+ DCs ( BALB/c ) did not produce significant IL-12 levels in the absence of parasites ( Figure 7C ) ., In contrast , IL-12 production was drastically increased upon addition of exogenous BK ( positive control ) or TCT , whereas HOE-140 cancelled both stimuli ( Figure 7C ) ., Notably , TCT induced IL-12-producing DCs irrespective of the presence/absence of lisinopril , a rather selective inhibitor angiotensin-converting enzyme ( ACEi ) ( Figure 7 ) ., Specificity controls confirmed that HOE-140 did not interfere at all with the magnitude of IL-12 responses induced by lipopolysaccharide ( LPS ) ( Figure 7C ) ., In agreement with the FACS data , ELISA determinations of IL-12 levels in cultures supplemented with HOE-140 confirmed that TCT activate immature DCs through B2R ( Figure 7D ) ., Controls in the absence of pathogen indicated that lisinopril or HOE-140 as such did not induce IL-12 production by DCs ( Figure 7C ) ., Additionally , DCs cultivated with either TCT or BK ( positive control ) displayed increased surface expression of CD40 and CD86 ( Figure 7E ) ., Of note , HOE-140 cancelled the phenotypic changes induced by TCT ( Figure 7E , upper and lower panels ) , while responses induced by BK were significantly reduced by this B2R antagonist ( Figure 7E , lower panel ) ., Since TCT generate kinins via CZP while invading endothelial cells , we next asked if parasite cysteine proteases were required for DC activation ., This question was addressed by pre-incubating TCT with methylpiperazine-Phe-homoPhe-vinylsulfone-benzene ( VSPh ) , an irreversible inhibitor of CZP ., After washing the VSPh-TCT , they were added to DC cultures ., Whether using FACS and ELISA , we found that VSPh-TCT failed to drive significant IL-12 production by DCs ( Figure 7C and 7D ) , adding weight to the concept that the parasite relies on CZP to generate the innate kinin stimuli ., In order to verify whether the B2R−/− CD11c+ DCs were fully capable of responding to TLR agonists , we compared the in vitro response profile induced by cytosine-phosphate-guanine ( CpG ) and LPS ., As shown in Figure 7F , IL-12 responses were of the same magnitude as compared to wild-type C57BL/6 DCs ., Moreover , HOE-140 did not interfere with wild-type DC responsiveness to CpG and LPS ( Figure 7F ) ., Notably , the magnitude of B2R−/− DC response to TCT was nearly 10% of IL-12 responses observed in wild-type CD11c+ DCs ( Figure 7F ) ., As expected , TCT or BK elicited vigorous IL-12 production in CD11c+ DCs from wild-type mice ., In both cases , the IL-12 response was partially blocked by HOE-140 ( Figure 7F ) ., In contrast , BK did not induce IL-12 in B2R−/− DCs ( Figure 7F ) ., As mentioned earlier , we found that production of IFN-γ by Ag-experienced CD4+ and CD8+ T cells from B2R−/− spleen and heart declined sharply as the infection continued ( 28 d p . i . ) ., In view of those findings , we asked whether the deficient type-1 responses of B2R−/− mice were restored upon adoptive transfer of wild-type DCs ., To address this question , we adoptively transferred ( intravenously ) immature B2R+/+ CD11c+ DCs ( 106 cells ) into B2R−/− mice before injection of the parasites ., As controls , recipient B2R−/− mice received an equivalent number of CD11c+ DCs isolated from donor B2R−/− spleen ., As expected , our controls showed that B2R−/− mice succumbed ( 100% mortality , n = 5; three independent experiments ) at day 30 ., In contrast , 100% of the B2R−/− recipient mice reconstituted with B2R+/+ DCs survived the acute challenge ., Of note , the mice of the specificity control group ( B2R−/− DCs → B2R−/− mice ) succumbed ( 100% ) to the infection , thus ruling out the possibility that adaptive immune function was restored due to non-specific activation of these APCs during the DC isolation procedure ., We then ran another set of experiments to verify if the DC transfer maneuver had restored ( type-1 ) acquired immunity of B2R−/− recipient mice ., Recall assays performed at day 28 p . i . confirmed that splenic or intracardiac ( CD4+ or CD8+ ) T cells from control B2R−/− mice secreted lower levels of IFN-γ as compared to experienced CD4+ or CD8+ T cells isolated from B2R+/+ spleen or heart ( Figure 8A ) ., Notably , B2R−/− mice that received adoptive transfer of B2R+/+ DCs recovered the ability to generate IFN-γ-producing CD4+ and CD8+ T cells ( Figure 8A ) ., Conversely , the DC transfer to B2R−/− mice repressed the secretion of IL-17 ( Figure 8B ) and TNF-α ( Figure 8C ) by Ag-experienced ( splenic or intracardiac ) CD4+ T cells of the reconstituted B2R−/− mice , therefore simulating the phenotype of wild-type-infected mice ., In the present work , we have demonstrated that the immune dysfunction of B2R−/− mice infected intraperitoneally with T . cruzi is a consequence of defective sensing of endogenously released kinins by immature CD11c+ DCs ., Our analysis of the adaptive immune responses of infected B2R−/− appointed a role for the kinin signaling pathway in the development of type-1 effector T cells ., The critical importance of DCs as sensors of kinins was confirmed by adoptive cell transfers ( wild type DC→ B2R−/− mice ) , which reversed the susceptible phenotype of B2R−/− mice while restoring the development of type-1 effector T cells , both in the spleen and cardiac tissues of recipient B2R−/− mice ., The notion that the kinin-releasing trypomastigotes induce DC maturation through B2R is supported by the following experimental evidence ., First , our in vitro studies showed that TCT vigorously induced IL-12 responses in splenic DCs originating from wild-type ( C57BL/6 ) mice , while failing to activate B2R−/− DCs ., Second , we demonstrated that HOE-140 , a specific antagonist of B2R , efficiently blocked DC maturation ( IL-12 induction , upregulation of CD80 , CD86 , and CD40 ) ., Furthermore , the irreversible inhibitor of CZP ( K11777 ) mitigated the IL-12 stimulatory activity ( B2R-driven ) of TCT , thus implicating the major cysteine protease of T . cruzi in the kinin generation mechanism ., Extending these observations to the in vivo settings , we then analyzed IL-12 production by splenic CD11c+ DCs isolated 18 h after systemic inoculation ( intravenously ) of Dm28c TCT ., Experiments performed with BALB/c mice showed that mice pre-treated with HOE-140 presented reduced frequencies of splenic CD11c+ IL-12+ DCs ., Adding weight to these results , we demonstrated that TCT induced high frequencies of CD11c+ IL-12+ DCs in wild-type ( C57BL/6 ) spleen , while failing to evoke significant IL-12 responses in DCs isolated from B2R−/− spleen ., Notably , preliminary studies indicated that macrophages ( CD11b+F4/80+ ) isolated from the spleen of these wild-type and B2R−/− mice develop comparable IL-12 responses ., Given that type-1 immune responses in the spleen of B2R−/− mice are well preserved at day 14 p . i . , it is possible that macrophages activated by alternative routes provide the IL-12 signals that drive adaptive immunity in this secondary lymphoid tissue ., Although we cannot claim that conventional DCs are the primary or even unique in vivo targets of T . cruzi in the spleen , the above mentioned results support the concept that kinin-releasing pathogens may drive DC maturation in vivo through the activation of G-protein-coupled B2 receptors 32 ., Since lymphoid tissues are irrigated by non-fenestrated capillaries , we may predict that trypomastigotes invading the splenic stroma are faced with an abundant supply of blood-borne proteins , such as kininogens ., Given biochemical evidence that interactions of high molecular weight kininogens with heparan sulfate proteoglycans potentiate the kinin-releasing activity of CZP 40 , it is plausible that the extracellular trypomastigotes might promptly liberate these paracrine signaling peptides while moving through extracellular matrices , hence driving DC maturation via B2R 31 , 32 ., At first sight , our finding that TCT induce DC maturation via the endogenous kinin/B2R pathway appears to conflict with the well-established concept that innate sentinel cells sense pathogens via pattern recognition receptors ( PRRs ) , such as the members of the TLR family 28 , 44 ., Indeed , early studies of macrophage ( IFN-γ-primed ) interaction with T . cruzi ( Y strain ) suggested that TLR2 and TLR4 ligands 20–22 are major drivers of innate responses in T . cruzi infection ., In a limited attempt to investigate the functional relationship of B2R and TLRs , we examined the outcome of TCT interaction in vitro with CD11c+ DCs ( splenic origin ) derived from either TLR2−/− or TLR4d/d mice ., Our results indicated that TCT induced vigorous IL-12 responses both in TLR2−/− DCs and TLR4d/d DCs ( unpublished data ) ., Moreover , we found that addition of HOE-140 to the TCT/DC culture system blocked IL-12 responses by TLR2−/− or TLR4d/d DCs ( unpublished data ) ., Admittedly , complementary studies with DCs from double TLR2/TLR4 knockout mice and MyD88−/− mice are required to rule out the possibility that B2R-responsive phenotypes of TLR2−/− DCs and TLR4d/d DCs reflect compensatory responses , respectively induced by TLR4 and TLR2 ligands of T . cruzi 20–22 ., The intertwined nature of the innate pathways controlling IL-12 production by APCs is illustrated by the recent demonstration 23 that T . cruzi DNA potently induces IL-12 production by mouse macrophages through the activation of TLR9 ., Given the evidence that DCs are parasitized by T . cruzi 45 , it will be interesting to determine if endogenous ( BK/LBK ) and exogenous ( T . cruzi DNA ) danger signals may activate their respective sensor receptors , B2R and TLR9 , at distinct temporal stages ( i . e . , early and late ) of intracellular infection ., While examining the frequencies of type-1 effectors in extra-lymphoid and lymphoid tissues of wild-type and B2R−/−-infected mice , we became aware that B2R deficiency affected the temporal and spatial distribution of IFN-γ-producing CD4+ and CD8+ T cells ., Recall assays performed at day 14 p . i . revealed weakened IFN-γ production by intracardiac CD4+ and CD8+ T cells isolated from B2R−/− mice ., However , we found high and comparable frequencies of INF-γ-producing T cells in the spleen of the same B2R−/− and wild-type mice ., Since the parasites are scarcely found in the spleens of wild-type and B2R−/− mice , we may infer that activation of the kinin/B2R pathway is dispensable for early induction of type-1 effectors in the spleen ., Adoptive cell transfer studies are required to find out if the induction of these early type-1 effector T cells is controlled by MyD88-coupled pathways 24 , such as those triggered by TLR2/TLR9 23 and/or by IL-1R/IL-18 R 44 ., In addition , it is possible that IL-12 induction by the NKT/CD1 pathway 29 may also contribute to early development of type-1 effectors in lymphoid tissues ., It is intriguing that intracardiac CD4+ and CD8+ T cells from B2R−/− mice ( 14 d p . i . ) showed impaired production of IFN-γ , despite the fact that the spleen of these mice displayed high frequencies of type-1 effectors ., Coincidently , tissue parasite burden is drastically increased in B2R−/− heart , thus showing an inverse correlation between these two parameters at day 14 p . i . Although we cannot a priori assume that Ag specificities of T cells recruited to the heart of wild-type and B2R−/− mice at 14 d p . i . are necessarily the same , independent studies performed with the Brazil 46 and Y strain of T . cruzi 47 converged in appointing cytotoxic CD8+ T cells as the key effectors controlling intracellular parasite outgrowth in cardiac tissues ., So far , efforts to characterize the Ag specificity of intracardiac CD8+ T cells in our infection model have been hampered by the findings that Dm28c T . cruzi strain did not present open reading frames for genes coding for ASP-2 antigens 48 , which in other systems provide dominant epitopes recognized by cytotoxic CD8+ T cells 46 , 47 ., In spite of these limitations , it is conceivable that immunoregulatory dysfunctions were responsible for the weakened type-1 responses observed in peripheral T cells from B2R−/− mice ., For example , it is possible that the migratory competence of effector T cells generated in lymphoid tissues may depend on DC activation via the kinin/B2R pathway ., Pertinently , recent analysis of the susceptible phenotype of CCR5−/− mice infected with T . cruzi implicated this chemokine receptor in the recruitment of CD8+ and CD4+ effector T cells into infected heart 13 , 14 ., Given these precedent findings , it will be worthwhile investigating if B2R and CCR5 signaling , whether acting separately or in conjunction , might promote the migration of effector T cells to peripheral sites of infection , such as the heart ., As the infection advanced ( 14→28 d ) , wild-type mice developed high frequencies of IFN-γ-producing CD4+ and CD8+ effector T cells , both in the spleen and heart ., Interestingly , a significant proportion of these Ag-responsive T cells displayed activated ( CD44+ ) phenotypes ., In contrast , B2R−/− mice showed negligible frequencies of activated type-1 effectors at day 28 , both in spleen and heart ., Of note , we found that the intracardiac CD4+ and CD8+ T populations recovered from the CD3+ pool of B2R−/− mice were significantly contracted ( Figure 5 ) ., Considering that B2R−/− mice recovered the capacity to mount protective type-1 responses upon adoptive transfer of wild-type DCs , it is possible that maintenance of T cell homeostasis may depend , at least to some degree , on DC responses elicited by endogenously released kinins ., Albeit speculative , this hypothesis is worth exploring in light of independent reports showing that aberrant T cell apoptosis is the primary cause of the immunoregulatory abnormalities underlying host susceptibility to acute infection by the Dm28c strain of T . cruzi 49 ., Another intriguing phenotypic characteristic of infected B2R−/− mice emerged when we monitored production of IL-17 and TNF-α in our recall assays ., Unexpectedly , we found that the weakened TH1 responses of B2R−/− CD4+ T cells ( whether isolated from the spleen/heart ) at day 28 d p . i . was accompanied by upregulated production of IL-17 and TNF-α , two pro-inflammatory cytokines associated with the effector activity of TH17 cells ., Recently characterized as a separate lineage of pro-inflammatory T helper cells distinct from conventional TH1 and TH2 cells 42 , 43 , TH17 cells differentiate from naïve precursors under the critical influence of IL-6 and TGF-β1 50 ., It is also known that committed TH17 cells depend on the IL-23 survival signal to develop their pro-inflammatory function in vivo 51 ., Notably , at early stages of infection ( 14 d p . i . ) , there was no significant production of IL-17 and TNF-α by spleen- or heart-derived T cells from infected B2R−/− mice , whether detected by conventional recall assays or polyclonal activation with anti-CD3 antibodies ( unpublished data ) ., It is unclear why the TH1/TH17 balance was inverted as the acute infection progressed in B2R−/− mice ., Recently , IL-27 was identified as the cytokine that suppresses TH17 differentiation driven by IL-6 and TGF-β via STAT-1 , independently of IFN-γ 50 ., Interestingly , T . cruzi–infected WSX-1 mice ( deficient in the IL-27Ra ) 52 develop severe hepatic injury , correlating with overproduction of various pro-inflammatory cytokines , such as IL-6 , TNF-α , and IFN-γ 52 ., Although TH17 responses were not evaluated in T . cruzi–infected WSX-1 mice , these animals strongly upregulated TH2 cytokines 52 ., However , we were unable to detect IL-4 production or IgG isotype switching in infected B2R−/− mice , indicating that these mice strains do not share the same phenotype ., Importantly , the recovery of type-1 responses in DC recipient B2R−/− mice was associated with reduced production of IL-17 and TNF-α ., Additional studies are underway to determine if DCs activated by the kinin/B2R pathway may influence TH1/TH17 lineage development in T . cruzi infection via IL-27 , or through alternative mechanisms ., Collectively , our results have linked development of acquired

Introduction, Results, Discussion, Materials and Methods

Although the concept that dendritic cells ( DCs ) recognize pathogens through the engagement of Toll-like receptors is widely accepted , we recently suggested that immature DCs might sense kinin-releasing strains of Trypanosoma cruzi through the triggering of G-protein-coupled bradykinin B2 receptors ( B2R ) ., Here we report that C57BL/6 . B2R−/− mice infected intraperitoneally with T . cruzi display higher parasitemia and mortality rates as compared to B2R+/+ mice ., qRT-PCR revealed a 5-fold increase in T . cruzi DNA ( 14 d post-infection p . i . ) in B2R−/− heart , while spleen parasitism was negligible in both mice strains ., Analysis of recall responses ( 14 d p . i . ) showed high and comparable frequencies of IFN-γ-producing CD4+ and CD8+ T cells in the spleen of B2R−/− and wild-type mice ., However , production of IFN-γ by effector T cells isolated from B2R−/− heart was significantly reduced as compared with wild-type mice ., As the infection continued , wild-type mice presented IFN-γ-producing ( CD4+CD44+ and CD8+CD44+ ) T cells both in the spleen and heart while B2R−/− mice showed negligible frequencies of such activated T cells ., Furthermore , the collapse of type-1 immune responses in B2R−/− mice was linked to upregulated secretion of IL-17 and TNF-α by antigen-responsive CD4+ T cells ., In vitro analysis of tissue culture trypomastigote interaction with splenic CD11c+ DCs indicated that DC maturation ( IL-12 , CD40 , and CD86 ) is controlled by the kinin/B2R pathway ., Further , systemic injection of trypomastigotes induced IL-12 production by CD11c+ DCs isolated from B2R+/+ spleen , but not by DCs from B2R−/− mice ., Notably , adoptive transfer of B2R+/+ CD11c+ DCs ( intravenously ) into B2R−/− mice rendered them resistant to acute challenge , rescued development of type-1 immunity , and repressed TH17 responses ., Collectively , our results demonstrate that activation of B2R , a DC sensor of endogenous maturation signals , is critically required for development of acquired resistance to T . cruzi infection .

Antibodies and IFN-γ-producing effector T cells are essential for the immune control of infection by Trypanosoma cruzi , the intracellular protozoa that causes human Chagas disease ., Despite the potency of anti-parasite immunity , the parasites are not cleared from their intracellular niches ., Instead , a low grade chronic infection prevails , provoking severe immunopathology in the myocardium ., Although it is well established that innate sentinel cells sense T . cruzi through receptors for microbial structures , such as Toll-like receptors , it remained unclear whether endogenous inflammatory signals also contribute to the development of adaptive immunity ., The present study was motivated by awareness that T . cruzi trypomastigotes ( extracellular infective forms ) are equipped with proteases that liberate the pro-inflammatory bradykinin peptide from an internal segment of kininogens ., Here we demonstrate that splenic dendritic cells ( DCs ) , the antigen-presenting cells that coordinate the adaptive branch of immunity in lymphoid tissues , are potently activated via G-protein-coupled bradykinin B2 receptors ( B2R ) ., Analysis of the outcome of infection in B2R-knockout mice revealed that the mutant mice developed a typical susceptible phenotype , owing to impaired development of IFN-γ-producing effector T cells ., Notably , the immune dysfunction of B2R-knockout mice was corrected upon cell transfer of wild-type DCs , thus linking development of protective T cells to DCs sensing of endogenous danger signals ( kinins ) released by trypomastigotes .

t. cruzi, infectious diseases, immunology, eukaryotes

journal.pcbi.1004438

Shaping Neuronal Network Activity by Presynaptic Mechanisms

Oscillatory activity patterns in the brain have been linked to sleep , sensorimotor gating , short-term memory storage and selective attention 1 , 2 ., Neuronal microcircuits in the brain spontaneously generate oscillatory activity patterns via synaptic interaction between groups of neurons 1 , 2 ., Indeed , changes in synaptic transmission cause alterations in neuronal firing and neuronal network activity 3–5 , and synaptic dysfunction can lead to pathological epileptic conditions 6–8 ., Even though small alterations in synaptic transmission and in the firing properties of single neurons can alter the spontaneous and evoked activity of entire neuronal circuits 3 , 9 , most computational models of neuronal networks do not explicitly account for the elaborate presynaptic neurotransmission process ., Presynaptic transmission is a regulated multistep process that encompasses the loading of neurotransmitters into synaptic vesicles , the translocation to and docking of those vesicles at the plasma membrane ( PM ) , and vesicle preparation for fusion through a calcium-dependent maturation process generally referred to as vesicle priming 10–14 ., This pool of primed vesicles is the readily releasable pool ( RRP ) , where vesicles undergo immediate fusion with the PM upon acute elevation in intracellular calcium concentration ( Ca2+i ) ., Another presynaptic pool of vesicles , the recycling pool ( ReP ) , accommodates unprimed vesicles which can undergo maturation and fusion during repetitive synaptic stimulation; all of the remaining vesicles in the presynaptic terminal belong to the reserve pool ( RP ) ., Equilibrium of the presynaptic vesicles transition between these pools depends on neuronal activity , synaptic proteins and calcium 15–19 ., In the synapses , there are three types of synaptic release modes that rely on the high dynamic range of Ca2+i and share the same vesicle pools 20 , 21 ( but see 22 , 23 ) ., They are defined by their temporal association with the action potential ( AP ) :, a ) synchronous release , driven by a short-lived acute increase in Ca2+i , is time-locked to the AP 24–26;, b ) asynchronous release begins several milliseconds after an AP and drives slower vesicle release; this rate is two orders of magnitude slower than that of synchronous release ., Asynchronous release is enhanced by slow clearance of residual calcium from the presynaptic terminal , as well as by strontium application 24;, c ) spontaneous release which emerges without any association to previous neuronal activity ., Although presynaptic transmission is well understood at the single-neuron level , it is unclear how the aforedescribed manipulation of presynaptic processes modulates the activity patterns and synchronization of the network ., Recently , a handful of studies have begun to investigate how manipulations of non-synchronous presynaptic release , such as asynchronous or spontaneous release , modulate neuronal network activity 3 , 6–8 , 24 , 25 , 27–30 ., Understanding the determinant properties of spontaneous activity of the neuronal network is highly complex ., Therefore , neuronal network computational models are utilized to simulate key features of the networks spontaneous activity ., A large group of simulations utilizes computationally light leaky integrate-and-fire ( LIF ) neurons to model the activity of large-scale neuronal networks 31 , 32 ., However , these neuronal models are based on very general assumptions regarding neuronal synaptic transmission processes and thus do not simulate critical synaptic mechanisms , such as the transition of vesicles between pools , vesicle maturation steps or calcium-dependent presynaptic release ., An important model for neuronal networks , which combines the concept of synaptic resources and neuronal activity , is the tri-state model 33 ., The original model , based on three kinetic equations , organized synaptic resources into three states: active , recovered or inactive ., Synaptic transmission in this model was determined by the available synaptic resources ( i . e . vesicles ) and a constant utilization factor ( i . e . calcium , according to the calcium-based synaptic release theory ) ., This model was later extended to include an increase in the utilization factor as the neuron keeps firing 34–36 , much like the increase in Ca2+i occurring in short-term synaptic plasticity ., Another extension of the model also included asynchronous synaptic transmission by adding a super-inactive state 29 , 30 , 37 , 38 to generate reverberatory activity in small networks ., Nonetheless , this model does not directly simulate the presynaptic vesicle pools , calcium-dependent vesicle priming or calcium-dependent release , which are basic and crucial properties of presynaptic release 25 , 39–42 ., Furthermore , in response to evoked stimulations , this model generates very short network oscillations ( each oscillation lasting several milliseconds ) , which are several orders of magnitude shorter and more frequent than the network bursts recorded in vitro ( typically several hundreds of milliseconds of recurrent network activity ) 43 , 44 ., In general , none of these models simulate spontaneous release , which is physiologically important 45–47 , and spontaneous activity in these models is generally achieved by artificial injection of current 48–50 ., In addition , a model that relates in detail to changes in synaptic processes , and provides a mechanistic explanation and prediction for how changes in synaptic mechanisms at the neuronal level govern the activity patterns and synchronization at the network level is lacking ., In this paper , we present a novel computational model that demonstrates how changes in synaptic transmission modulate neuronal network activity patterns ., We utilized experimental data from in vitro neuronal networks cultured on microelectrode arrays ( MEA ) that spontaneously generate network-wide synchronized activity patterns , termed network bursts ., We used the model to learn about synaptic mechanisms that can explain changes in neuronal network activity following manipulations of the presynaptic release processes 3 , 5 , 43 , 44 , 51 ., Our model attempts to strike a balance between detailed cellular models and simplified neuronal network models 15 , 21 , 26 , 52 by extending the LIF neuronal model to simulate both the presynaptic release process and the entire neuronal network ., This allowed us to examine how manipulations of specific steps in the presynaptic release mechanism affect neuronal network activity ., The model provides putative mechanistic explanations for various network activity patterns in vitro , such as network burst termination , and allows us to predict how changes in the presynaptic release machinery will affect network oscillation frequency ., We previously explored 3 how genetic and pharmacological manipulations of presynaptic release change the spontaneous activity of neuronal networks cultured on MEA plates ( Figs 1A and S1 ) ., To do so , we genetically and pharmacologically manipulated different synaptic transmission steps in cultured neuronal networks and examined the effects on neuronal network activity ., Pharmacological enhancement of asynchronous release was achieved by strontium application , which has been shown to activate calcium-dependent release mechanisms but is cleared from the presynaptic terminal more slowly than calcium 53 , 54 ., Genetic manipulations consisted of overexpressing DOC2B , a presynaptic protein that enhances spontaneous and asynchronous neurotransmitter release 55–57 ., Our findings suggested that higher levels of asynchronous release at single synapses , induced by DOC2B overexpression or by strontium application , increase the firing rate within a network burst; on the other hand , facilitation of spontaneous release frequency by overexpression of the DOC2BD218 , 220N mutant 3 led to lower network burst firing rate ( Figs 1B and S1 ) ., These findings join other studies that have shown that manipulation of presynaptic proteins has a substantial impact on neuronal network plasticity , information transfer and animal behavior 10 , 58 , 59 ., However , it is difficult to infer a mechanistic explanation for these findings ., Therefore , we developed a computational model that simulates how changes in different steps of synaptic transmission alter neuronal firing ., The model consisted of 800 LIF neurons , spread on a virtual MEA-like 2D surface ( 30% inhibitory neurons; Fig 1C ) 3 ., The neurons were connected by the small-world and scale-free topology typically associated with cortical neuronal networks 60–63 ( S2 Fig ) , creating an active neuronal network ., A key feature of the model was that neuronal activity and synaptic release were generated from a presynaptic compartment that simulates the multistep process of calcium-dependent synaptic transmission ., This presynaptic compartment was simulated for each LIF neuron ( Eq, 1 ) and governed the spontaneous , evoked and asynchronous activity of each neuron in the network ., All of the chosen parameters were based on up-to-date papers ( Table, 1 ) 63 ., Our model allowed us to perform in silico experiments , manipulate specific properties of synaptic transmission and study their impact at the network level ., It gave us access to multiple cellular parameters , such as vesicular pool capacities , vesicle replenishment rate and Ca2+i , and simultaneously follow the macroscale network activity and the interaction between neurons ., Each neuron received multiple inputs which accumulated as changes in the PM voltage until they crossed a threshold ( Eq, 2 ) and generated an AP or decayed with a predefined time constant ( Table 1 ) ., AP generation induced a transient increase in the Ca2+i that accumulates when several APs arrive concomitantly ( Eq 4 ) ., This increase in calcium was then translated into vesicle release according to a calcium-dependent synaptic release curve ( Fig 1D ) ., The release curve ( described in Eq 5 ) linked the free synaptic Ca2+i to synaptic release probability ( Pr ) according to well-established release-rate curves 21 , 25 , 26 ., According to most calcium-dependent release models , upon AP generation , calcium level increases by almost two to four orders of magnitude in the active zone , inducing an acute shift in the synaptic Pr 25 , 26 ., Accordingly , we used the Calyx of Held calcium-dependent release-rate curve as previously described 26 with a small modification to fit the lower Pr of cortical synapses ., To recreate the multiscale temporal dynamics of synaptic release , each synapse consisted of three vesicle pools: RP ( 170 vesicles ) , ReP ( 20 vesicles ) and RRP ( 10 vesicles ) ( Fig 1E ) ; the vesicle transportation between pools was bidirectional ( Eqs 7 and 8 ) ., Following vesicle release , vesicles underwent refilling according to different rate constants ( Table 1 ) ., A variety of neuronal preparations have demonstrated that vesicle recruitment in neurons is enhanced by elevated Ca2+i 41 , 64 , 65 , and this enhancement has been recognized as essential for maintaining adequate release during high-frequency bursts of activity 65 , 66 ., Therefore , we adapted the rate of vesicle transition from the ReP to the RRP to a similar Michaelis–Menten-type equation ( Fig 1E black frame; Eq 7 ) which has been used to describe the calcium-dependent transition rate from the unprimed pool to the RRP in chromaffin cells 15 , 67 ., Each vesicle fusion event contributes a positive or negative voltage upon release ( excitatory or inhibitory postsynaptic potential , respectively ) to the PM of the postsynaptic neuron ., Notably , the basal activity in the model was maintained by spontaneous release driven from the Pr of the neuron under resting calcium levels ( Eq 5 ) ., This method kept the network active and replaced the common route of keeping computational neuronal networks spontaneously active , i . e . injecting current into the neurons 48–50 ., Comparison of network spontaneous activity between these two methods showed that calcium-dependent synaptic release generates network bursts which are more similar to those recorded from neuronal networks cultured on MEA ( S3 Fig ) ., Recurrent network-wide bursting activity and abundant inter-burst activity can be observed in the color-coded raster plot of neuronal network spontaneous activity generated by the model ( Fig 1F ) ., Hence , the model recreated a pattern of synchronized activity followed by a period of quiescence similar to that in the experimental recordings ( compare Fig 1F to 1A ) ., Importantly , the model recreated both network-wide bursting activity ( full bursts; green box ) and bursting activity limited to subnetworks ( aborted bursts; black box ) ., To test the stability and robustness of the network activity under various manipulations , we explored the response of the model to changes in its primary gain parameters: excitatory postsynaptic potential ( EPSP , voltage ) and connectivity ratio ( the percentage of actual connections out of all possible connections in the network; see considerations for choosing these parameters in Methods ) ., Quantitative analysis of the basic model activity parameters , such as global and network burst spike rate , network burst frequency and network burst duration , was performed under different levels of the gain parameters ( S4 Fig ) ., We found that the model is robust to two- to threefold changes in basic gain parameters while maintaining continuous spontaneous network activity but displaying changes in various network activity properties ( S4 Fig ) ., We also showed that even increasing the number of neurons or the number of synapses in the model 10-fold does not change its basic bursting activity; the neuronal network still displayed network-wide bursts followed by periods of relative quiescence ( S5 Fig ) ., The stability of the bursting activity of the network following changes in basic gain parameters ( and changes in the number of neurons and number of independent synapses per neuron ) established the robustness of the model and increased its fidelity ., Indeed , most of the experimental manipulations did not abolish the basic bursting activity in the network but rather manipulated the inter-burst and intra-burst spiking profiles ., This places the model in an excellent position to test the impact of changes in other parameters of synaptic release on the network bursting activity ., The established model was utilized to understand two intriguing findings: elevated asynchronous release , but not spontaneous release , at the single-neuron level enhances and synchronizes network burst activity 3; on the other hand , enhanced spontaneous release reduces synchronization and network burst activity ., Experimentally , asynchronous release was elevated by either DOC2B or strontium ., Strontium has been suggested to trigger vesicle fusion and neurotransmitter release in the same way as calcium , but is extruded from the synapse more slowly than calcium , causing long-lasting vesicle fusion or asynchronous neurotransmitter release 53 , 68 ., Therefore , to mimic the effect of asynchronous release , we reduced the rate of calcium efflux out of the synapse ( Eq 4 , τCafast and τCaslow ) , allowing more time for vesicle fusion 69 ., It is important to note that we changed the asynchronous release in both excitatory and inhibitory neurons ., We first verified that slower calcium clearance increases the ratio of asynchronous to synchronous release in the model ., We followed the change in the probability for vesicle release from single neurons up to 50 ms after an AP , under different calcium-efflux rates ( Fig 2A; see Methods ) ., The ratio of asynchronous to synchronous release ( Fig 2A right panel; ASync and Sync , correspondingly ) increased as calcium efflux was reduced 70 ., Notably , the increase in asynchronous release did not increase the total neuronal output of a single neuron but only spread the release over a longer time ., We then examined how asynchronous release affects the activity profile in the network burst ( Fig 2B ) ., Gradually increasing asynchronous release in the model enhanced the network burst firing rate ( Fig 2B; +100% , left panel ) , similar to the experimental results of increasing strontium concentration ( Fig 2B right panel ) ., Both manipulations also decreased the time from burst onset to its peak ., Interestingly , even when we increased the number of neurons in the network 10-fold ( 8000 instead of 800 ) and also when we increased the number of synapses per neuron 10-fold ( 10 instead of 1 ) , enhanced asynchronous release facilitated network burst firing rate and decreased the network bursts time to peak ( S5 Fig ) ., This supports the power of the model in mimicking experimental results and suggests that asynchronous release has a profound effect on neuronal network activity ., Next , we focused our analysis on the manipulation of spontaneous release and tested its effects on the network activity ., Experimentally , spontaneous release was increased by overexpressing a DOC2B mutant , DOC2BD218 , 220N , that is known to increase spontaneous release 55 ., Computationally , spontaneous release was elevated by increasing the Pr at resting calcium ( Fig 2C top panel; Eq 5 ) ., This manipulation increases the probability of vesicle release under resting conditions , which is the basic definition of spontaneous release 25 ., The increase in spontaneous release in the model led to a significant decrease in the network burst activity , as evidenced by the reduced network burst activity profile and the lower global spike rate in each network burst ( Fig 2C and 2D ) ., This manipulation recreated the experimental data of DOC2BD218 , 220N overexpression ( Fig 2C; compare bottom left panel , model , to bottom right panel , experiment ) while reducing the number of spikes and the number of neurons in the network bursts ( Fig 2D ) ., Comparison of the changes induced by both manipulations established their opposite effects on network activity ( Fig 2D ) ; while asynchronous release was positively correlated with network burst activity , spontaneous release was anticorrelated ., This means that specific activity properties can change in the same direction by an increase in asynchronous release or a decrease in spontaneous release , or vice versa ., These opposite effects were more prominent in the global spiking rate and network burst spikes; however , the burst rate , for example , displayed a more prominent difference between spontaneous and asynchronous release upon an increase in the corresponding parameter ( Fig 2D ) ; while higher spontaneous release reduced network burst frequency , lower spontaneous release did not change it ( Fig 2D , bottom panel ) ., Therefore , it is important to examine the combination of various network activity parameters to determine the overall effect on the network activity ., Next , we examined whether the model recreates the higher-level effects on network activity patterns observed in the experimental results 3 ., Evidently , higher asynchronous release in the model significantly increased , while spontaneous release reduced the ratio of neurons participating in the network bursts ( S6 Fig ) ., This was measured by classifying network bursts into full or aborted bursts 3 , 44 ., Moreover , analysis of the normalized network burst synchronization in the simulation showed that elevated asynchronous release also increases network burst synchronization , primarily around the peak of the network burst ( S6 Fig ) ., These analyses were in agreement with the experimental findings and showed that the model successfully recreates the response to the manipulation of asynchronous and spontaneous release ., Thus , using the in silico model , we manipulated specific steps in the release process and linked them to specific experimental changes ., Hence , the model reaffirmed a wide range of experimental analyses , from basic firing rate to high-level network synchronization parameters ., The high reliability of the in silico model in reconstructing experimental findings allowed us to utilize it to explore the neuronal mechanisms underlying the findings and uncover the model parameters and factors that govern network activity ., Specifically , the model allowed us to follow neuronal parameters , such as changes in the various vesicle pools , which are unavailable experimentally ., We analyzed the vesicle pool dynamics and Ca2+i of the model neurons under baseline release levels ( Baseline ) and under enhanced asynchronous release ( +100% ) ., Fig 3A demonstrates changes in the number of RRP vesicles in 4 representative neurons throughout a single network burst ., Each neuron displayed different release patterns from the RRP but all displayed a certain degree of vesicle depletion ( Fig 3A , 3B and 3D ) ., Analysis of the average RRP occupancy in all neurons in all network bursts ( in 10 simulations ) showed that during the burst , the RRP are depleted by the same percentage under both baseline and enhanced asynchronous release conditions ( Fig 3B ) ., Further analysis of the average RRP content showed that most neurons have more than 4 vesicles in the RRP at the onset of the network burst and less than 2 vesicles at its termination ( out of a maximum occupancy of 10 vesicles in the RRP; Fig 3D ) ., It can be suggested that under these conditions , where more than 70% of the neurons have less than 2 vesicles left in the RRP ( i . e . less than 20% of the entire synaptic reservoir is available ) , network bursts are terminated ., This is not surprising but rather provides a clear connection between vesicle pool depletion and burst termination and a mechanistic explanation for previous experimental results 4 , 71 ., This analysis could not explain the increase in network activity under enhanced asynchronous release and therefore we continued to examine the changes in ReP dynamics , which transfers vesicles to the RRP through calcium-dependent vesicle priming ., The same analysis applied to the ReP showed that asynchronous release manipulation causes enhanced consumption and larger depletion of vesicles from this pool ( Fig 3C ) ; while only 7% of the neurons had less than 8 vesicles in the ReP at the time of network burst termination under baseline conditions , ~46% of the neurons had less than 8 vesicles under enhanced asynchronous release at the time of network burst termination ( Fig 3E ) ., On average , approximately 2–3 additional vesicles were consumed from the ReP during a network burst under enhanced asynchronous release ( an increase of 10–15% in total synaptic release , on average; Fig 3D ) ., This suggests that the ReP is the source for the higher output following elevated asynchronous release and that asynchronous release , driven by slower calcium clearance , relies on the replenishment rate of the ReP for support of the increased network activity ., To examine this hypothesis , we determined the average cumulative neuronal output throughout the burst ( Fig 3F ) ., On average , each neuron with a higher asynchronous release contributed ~2 more vesicles within the first 300 ms of the burst overall ., This accumulated increase underlies the higher network activity and synchronization during the bursts; it also supports our hypothesis that the ReP is the source vesicle pool contributing to this network effect ., The lower calcium efflux rate from the presynaptic terminal allows faster and larger accumulation of free calcium throughout the network burst ( Fig 3G ) ., This , in turn , has two important implications in the neuronal release dynamics throughout the burst:, 1 ) higher calcium levels lead to higher Pr;, 2 ) higher calcium levels increase the vesicle transition rate from ReP to RRP ( much like the calcium-dependent vesicle replenishment hypothesis ) ., Thus , the model revealed that the higher asynchronous release temporally increases the Pr and vesicle availability , causing enhanced neuronal network activity only during bursts ., Furthermore , this analysis pointed to the ReP as the vesicle pool that supports this increase in neuronal vesicle release and network synchronization ., Our model presented us with an opportunity to predict the effect of manipulations , which can be later examined experimentally ., We were therefore interested in testing how changes in priming rate at the single-neuron level affect network activity ., To implement this manipulation , we changed the maximum rate of vesicle transition from ReP to RRP ( Fig 4A , circled red marker; parameter Rmax in Fig 1E; Eq 7 τReP→RRP ) ., A comparison of raster plots showed that as the priming rate increases , the activity and frequency of the bursts are enhanced , while decreasing the priming rate reduced network activity ( Fig 4B ) ., Burst profile and activity parameter analyses supported these findings , suggesting that a 50% increase in priming rate would lead to ~30% increase in the maximum firing rate within the network burst ( Fig 4C and 4D ) and an increase of 4 bursts per minute in network burst frequency , i . e . the network displays a higher rate of oscillations following this manipulation without elevating the inter-burst activity ., Next , we overexpressed Munc13-1 , a positive regulator of vesicle priming rate , in neuronal networks plated on MEA , and found that a 2-fold increase in Munc13-1 expression levels increases the frequency of network bursts by 60% ( previously published in Lavi et al . 3; Fig 4E ) ., These results match the model predictions ( Fig 4B , High priming ) and suggest that changes in vesicle priming rate at the neuronal level tune the burst frequency at the network level ., Current computational network models do not simulate synaptic vesicle pools or calcium-dependent processes ., Many computational network models based on LIF neurons simulate neuronal activity as the sum of voltage or current input on the PM to recreate neuronal network activity ., Although this voltage accumulation causes the generation of an AP in the soma , it is the calcium influx through voltage-dependent calcium channels in the presynaptic terminal that drives the actual vesicle fusion and subsequent synaptic release 72 , 73 ., Therefore , free intracellular calcium dynamics gates the transfer of synaptic information from one neuron to the next , and the combination of calcium dynamics and vesicle release probability underlies short-term plasticity in the presynaptic terminal , a key mode of operation in several central synapses 41 , 74–78 ., Therefore , it is highly important to integrate calcium-dependent synaptic release , as we did in the current model , into LIF neuronal models ., Although the well-established tri-state model 33 did incorporate synaptic transmission into LIF neurons , that model and its succeeding extensions 29 , 30 , 34 , 36–38 did not simulate synaptic vesicle pools , calcium-dependent vesicle priming or vesicle release ., The evoked activity simulated in those models generated very short network oscillations ( several milliseconds ) , significantly shorter than the network bursts observed in vitro in dissociated neuronal cultures by single-neuron current-clamp recordings and neuronal network MEA recordings ( typically hundreds of milliseconds ) ., Furthermore , to maintain spontaneous network activity , current or voltage are artificially injected from an external source 48–50 ., The lack of biological mechanisms in the neuronal model makes it harder to infer physiologically relevant consequences and predictions ., The uniqueness of our model lies in its direct simulation of key components of the presynaptic release process , thereby revealing how changes in the release process , such as changes in release probability ( Eq 5 ) , vesicle pool size ( Eqs 7 and 8 ) , and calcium dependency ( Eqs 4–7 ) , affect neuronal network activity ., This direct simulation allows us to model synchronous , asynchronous and spontaneous release as derivatives of the same calcium-dependent release mechanism with different ranges of Ca2+i 25 , 26 , 41 ., The model also incorporates calcium-dependent vesicle priming ( Eq 6 ) , which is usually not modeled in neuronal network models ., Since our model directly simulates the presynaptic vesicle pools and calcium-dependent priming based on measured rate constants , the derivation of putative physiologically relevant mechanistic explanations from its predictions is more intuitive ., The model generates network bursts that are similar to those observed in MEA recordings ( S3 Fig ) in duration and firing rate , thereby enabling an investigation of mechanisms for burst termination under spontaneous neuronal activity , and linking them to the dynamics of vesicle pool depletion 79 ., Note that we are not claiming that it is impossible to create network bursts without incorporation of the presynaptic release mechanism , but rather that through the structure of our model , we were able to relate the network bursts to their underlying realistic and biologically plausible presynaptic mechanisms ., Our simulation allowed us to perform long-term in silico experiments ( which we limited to several hours ) , while the model clearly exhibited stability and robustness to changes in the primary gain parameters that control network activity—i . e . , EPSP and connectivity ratio ., In agreement with our experimental results , most of the manipulations performed in the model did not abolish the basic network bursting activity but rather manipulated the inter-burst and intra-burst spiking distributions ., The fact that the basic bursting activity of the model was not diminished after these manipulations establishes the models robustness and its provision of a stable platform to uncover the role of asynchronous and spontaneous release in neuronal network oscillatory activity ., A recent intriguing experimental finding demonstrated that asynchronous release , but not spontaneous release , enhances network activity and network burst synchronization 3 ., The model allowed us to test how changes in asynchronous release and spontaneous release affect network activity at the neuronal level ., Supported by experimental results , the model showed that higher spontaneous release leads to lower firing rate , lower neuronal participation in network bursts and lower frequency of bursts ., Higher spontaneous release reduced synchronization of the network activity by the superfluous release of vesicles throughout; this excess activity reduced the availability of releasable vesicles from the RRP during network bursts , which resulted in lower intra-network burst activity and synchronization ., The model also allowed testing whether the anticipated change following strontium application—enhanced asynchronous release—is translated into enhanced activity during the bursts , and investigating the vesicular source for this effect ., The model-simulated increase in asynchronous release elevated the overall activity of the network and various network burst parameters , including synchronization and neuron participation ., This was in agreement with the experimental results obtained following gradual application of strontium ., These results are supported by previous evidence regarding the link between asynchronous release and reverberatory activity 54 ., Previous studies have shown that rapid recovery of the RRP supports asynchronous release at the neuronal level 71 , and have suggested that network bursting activity depends on the vesicle depletion rate from the RRP 4 , 80 ., Here we suggest that during ongoing network activity , the neurons in the network are not fully depleted at the termination of the network burst ., Rather , examination of the RRP and ReP of all neurons in the network showed that under baseline conditions , it is sufficient that 70% of the neurons have less than 2 vesicles in the RRP ( that is , less than 20% of the overall vesicles available for immediate release in the RRP ) to termina

Introduction, Results, Discussion

Neuronal microcircuits generate oscillatory activity , which has been linked to basic functions such as sleep , learning and sensorimotor gating ., Although synaptic release processes are well known for their ability to shape the interaction between neurons in microcircuits , most computational models do not simulate the synaptic transmission process directly and hence cannot explain how changes in synaptic parameters alter neuronal network activity ., In this paper , we present a novel neuronal network model that incorporates presynaptic release mechanisms , such as vesicle pool dynamics and calcium-dependent release probability , to model the spontaneous activity of neuronal networks ., The model , which is based on modified leaky integrate-and-fire neurons , generates spontaneous network activity patterns , which are similar to experimental data and robust under changes in the models primary gain parameters such as excitatory postsynaptic potential and connectivity ratio ., Furthermore , it reliably recreates experimental findings and provides mechanistic explanations for data obtained from microelectrode array recordings , such as network burst termination and the effects of pharmacological and genetic manipulations ., The model demonstrates how elevated asynchronous release , but not spontaneous release , synchronizes neuronal network activity and reveals that asynchronous release enhances utilization of the recycling vesicle pool to induce the network effect ., The model further predicts a positive correlation between vesicle priming at the single-neuron level and burst frequency at the network level; this prediction is supported by experimental findings ., Thus , the model is utilized to reveal how synaptic release processes at the neuronal level govern activity patterns and synchronization at the network level .

The activity of neuronal networks underlies basic neural functions such as sleep , learning and sensorimotor gating ., Computational models of neuronal networks have been developed to capture the complexity of the network activity and predict how neuronal networks generate spontaneous activity ., However , most computational models do not simulate the intricate synaptic release process that governs the interaction between neurons and has been shown to significantly impact neuronal network activity and animal behavior , learning and memory ., Our paper demonstrates the importance of simulating the elaborate synaptic release process to understand how neuronal networks generate spontaneous activity and respond to manipulations of the release process ., The model provides mechanistic explanations and predictions for experimental pharmacological and genetic manipulations ., Thus , the model presents a novel computational platform to understand how mechanistic changes in the synaptic release process modulate network oscillatory activity that might impact basic neural functions .

journal.pcbi.1002122

A Mathematical Model for the Reciprocal Differentiation of T Helper 17 Cells and Induced Regulatory T Cells

CD4+ T cells are important components of the adaptive immune system in higher vertebrates ., By producing various cytokines , they perform critical functions such as helping B cells to produce antibodies , activating CD8+ cytotoxic T cells , enhancing the innate immune system , and suppressing the immune response to avoid autoimmunity 1 , 2 , 3 ., In peripheral tissues , such as lymph nodes , blood and sites of infection , antigen-inexperienced ( naïve ) CD4+ T cells can differentiate into effector cells of specialized phenotypes upon stimulation by cognate antigen delivered to the T cell receptor by Antigen Presenting Cells ( APCs ) ., Proliferation and differentiation of activated naïve T cells depends on their particular cytokine microenvironment ., These specialized effector T cells produce distinct cytokine profiles tailored for their specialized functions ., Also , they express lineage-defining transcription factors ( “master regulators” ) ., In general , high expression level of a particular master regulator is observed only in cells of a particular lineage , and the overexpression of a particular master regulator induces the production of the corresponding lineage-defining cytokines 4 , 5 ., The fate of a naïve CD4+ T cell was traditionally thought to be either T helper 1 ( TH1 ) cell or T helper 2 ( TH2 ) cell 6 ., In the last decade , a third type of T helper cell ( TH17 ) , derived from naïve CD4+ T cells , was discovered 7 ., TH17 cells produce interleukin-17A ( IL-17A ) , IL-17F and IL-22 as their lineage-defining cytokines , and the retinoic acid receptor-related orphan receptor gamma t ( RORγt ) transcription factor is considered the master regulator of this lineage 8 , 9 ., In addition , naïve CD4+ T cells were found to be able to differentiate into a fourth lineage of ( regulatory ) T cells , which were called induced regulatory T ( iTreg ) cells to distinguish them from natural regulatory T ( nTreg ) cells , which differentiate in the thymus instead of the periphery 10 ., iTreg cells are characterized by producing IL-10 and transforming growth factor-β ( TGF-β ) and highly expressing forkhead box P3 ( Foxp3 ) transcription factor as their master regulator 11 ., TH17 cells are pro-inflammatory because they secret cytokines that promote inflammation , whereas iTreg cells are anti-inflammatory because their lineage-defining cytokines can reduce the inflammatory response ., The differentiation pathways of naïve T cells into TH17 and iTreg lineages are closely related ., First , stimulation by TGF-β is necessary for the differentiation of both lineages 12 ., The differentiation of TH17 and iTreg cells are reciprocally regulated in the presence of TGF-β , i . e . inhibiting the differentiation pathway of one lineage will result in activation of the pathway for the other lineage ., This is due to the mutual antagonism between RORγt and Foxp3 ., Furthermore , polarizing signals , such as IL-6 and retinoic acid , can induce the differentiation of one lineage and repress that of the other one 12 ., Nonetheless , differentiated iTreg cells can be reprogrammed into TH17 cells in an appropriate cytokine environment 13 , suggesting significant plasticity of these two lineages ., In addition , stable co-expression of their master regulators ( RORγt and Foxp3 ) is observed both in vivo and in vitro 14 , 15 ., Interestingly , these double-expressing cells were found to possess either regulatory or dual ( regulatory and proinflammatory ) functions in vivo 14 , 15 ., Perhaps the most intriguing phenomenon is that antigen-activated naïve CD4+ T cells treated with TGF-β alone give rise to a heterogeneous population , which may include three phenotypes ( Foxp3-only , RORγt-only , and double-expressing cells ) at an intermediate TGF-β concentration 16 , or two phenotypes ( RORγt-only and double-expressing cells ) at a higher TGF-β concentration 15 ., In combination with TGF-β , IL-6 can induce the differentiation of RORγt expressing cells , whereas all-trans retinoic acid ( ATRA ) can induce the differentiation of Foxp3 expressing cells 16 , 17 ( Figure 1 ) ., All of these in vitro derived phenotypes can be observed in vivo , and at least some of their respective functions have been demonstrated , suggesting that these in vitro differentiation assays provide important clues to our understanding of the development of TH17 and iTreg cells in the body ., Mathematical modeling has contributed to our understanding of the differentiation of TH1 and TH2 cells 18 , 19 , 20 , 21 , 22 , 23 , 24 ., Höfer et al . first demonstrated that the dynamics of the key transcription factors can govern the robustness of the lineage choice and maintenance 18 , 19 ., Yates et al . later related transcription factor dynamics to the mix of TH1 and TH2 cells in a population of differentiating T cells 20 ., Recently , Bonneau et al . 25 have proposed a Boolean-network model of the comprehensive repertoire of CD4+ T cell phenotypes , including TH17 and iTreg cells ., Drawing inspiration from these earlier models , we have sought to explain , with a computational model , the remarkable heterogeneity of the TH17-iTreg reciprocal-differentiation system ., In terms of this model , we show that a population of naïve CD4+ T cells , with some small cell-to-cell variability , can differentiate into a heterogeneous population of effector cells with distinct phenotypes upon treatment with the primary differentiation signal ( TGF-β ) ., Polarizing signals , such as IL-6 and ATRA , can skew the differentiation to one or two phenotypes ., A control system with these properties can generate functional diversity of the induced cell populations and can be regulated with great flexibility by diverse environmental cues ., In addition , the model suggests how treatment with different concentrations of TGF-β may favor different responding phenotypes , and how conversions among these phenotypes may be guided ., Finally , the model gives a new quantitative explanation for double-expressing cells , suggesting that they are ‘re-stabilized co-expressing’ cells rather than transient intermediate cells in the differentiation pathway ., The model predicts that double-expressing cells should appear at a relatively late stage of the differentiation process , and they may be intended for specific functions ., In all , our model provides a novel mathematical framework for understanding this reciprocal differentiation system , and it gives new insights into the regulatory mechanisms that underlie the molecular control of certain immune responses ., To illustrate our basic idea , we first construct a model of a simple and perfectly symmetrical regulatory network ( Figure 2A ) ., In the Methods section we describe how this network is converted into a pair of nonlinear ordinary differential equations ( ODEs ) for the time rates of change of Foxp3 and RORγt ., The rate functions for this model contain 12 kinetic parameters , whose basal values are specified in the Methods section ( Table 1 ) for the “symmetrical model without intermediates” ., The solution of these ODEs for the basal values , and with TGF-β\u200a=\u200a0 , evolves to a stable steady state where both RORγt and Foxp3 have a low level of expression ( RORγtlowFoxp3low ) ., This steady state corresponds to a naïve CD4+ T cell ( Figure 3A ) ., In the presence of a sufficient TGF-β signal , the regulatory network might evolve to one of three other steady states , namely RORγthighFoxp3low , RORγtlowFoxp3high and RORγthighFoxp3high states , corresponding to RORγt-only , Foxp3-only and double-expressing phenotypes ., Note that these stable steady states are also referred as ‘cell fate attractors’ in some other studies , and this concept facilitates our understanding of cell lineage choice and reprogramming ( reviewed in 26 ) ., Figure 3B shows a scenario in which the TGF-β signal triggers the formation of a tri-stable system ., In this particular case , the RORγtlowFoxp3low state is no longer a stable steady state , and naïve cell , which was previously stabilized in the RORγtlowFoxp3low state , would differentiate into the RORγthighFoxp3high state , whose basin of attraction ( the white region in Figure 2B ) contains the naïve state of the cell ., However , cell-to-cell variability can produce other results ., We interpret cell-to-cell-variability as small deviations of the parameter values from their basal settings in Table 1 ., The basal settings correspond to the behavior of an “average” cell , but any particular cell will deviate somewhat from this average behavior ., As consequences of the changing parameter values in any particular cell , the position of the RORγtlowFoxp3low state changes , the boundaries of the basins of attractions change , and the fate of the naïve cell may change ., The naïve T cell will differentiate into the stable steady state in whose basin of attraction it lies ., That is , depending on the precise parameter values of the cell , its RORγtlowFoxp3low state may lie in any of the three basins of attraction of the TGF-β-stimulated system ., Figure 3C depicts three cells in the population that adopt three different fates because of the variability among them ., With a random sample of cells , each of the three differentiated states can be populated by a significant fraction of cells ( Figure 3D ) ., Although cell-to-cell variability does not make large changes in the position of the RORγtlowFoxp3low state , it has a dramatic influence on the basins of attraction of the stable steady states , which determines the fate of the cell once the differentiation signal is turned on ., Since the system has four distinct steady states that correspond to four distinct phenotypes , we next looked for the relationships among these steady states using bifurcation analysis of an average cell ., Because of the symmetrical nature of the interactions , an average cell exhibits sub-critical pitchfork bifurcations with TGF-β concentration as the control parameter ( Figure 4A ) ., ( The notion of a pitchfork bifurcation was used earlier , in references 27 , 28 , to explain a system of hematopoietic cell differentiation in which multiple lineages might be adopted . ), Notably , the RORγtlowFoxp3low state is only stable at low TGF-β concentration ., At an intermediate concentration of TGF-β ( ∼0 . 25 units in Figure 3A ) , the system bifurcates into two lineage-specific branches , corresponding to RORγthighFoxp3low and RORγtlowFoxp3high states ., The fourth type of stable steady state ( RORγthighFoxp3high ) appears at higher TGF-β signal strength ( >0 . 37 in Figure 3A ) , when the autoactivation of RORγt and Foxp3 eventually overrides their mutual inhibition and makes the double-expressing state the dominant phenotype of the population ., We next checked the influence of TGF-β concentration on the fractions of responding phenotypes in a population of induced cells ., For various values of TGF-β , we simulated a population of naïve CD4+ T cells with cell-to-cell variability ., In agreement with the bifurcation analysis , RORγthighFoxp3low and RORγtlowFoxp3high cells appeared simultaneously over an intermediate range of TGF-β ( between ∼0 . 2 and ∼0 . 55 units ) ., The fraction of RORγthighFoxp3high cells increases at higher TGF-β concentrations and eventually dominates the population when TGF-β>0 . 55 ., In the vicinity of 0 . 5 units of TGF-β , the cell population is heterogeneous , with comparable fractions of all three stable phenotypes ( Figure 4A lower panel ) ., Although this initial model accommodates the presence of dual-positive TH17/iTreg cells , it cannot fully explain the fine regulatory effects of varying TGF-β concentrations ., For example , this model predicts that double-expressing cells dominate the population when TGF-β concentration is high , and that single-expressing cells may be converted into double-expressing cells by increasing TGF-β ., In fact , this is not necessarily true if the effects of TGF-β saturate at high TGF-β ., To take saturation effects into account , we incorporated two intermediate signaling proteins between TGF-β and the transcription factors Foxp3 and RORγt ( Figure 2B ) ., In this case , the system can be tri-stable even at high concentrations of TGF-β , and the total conversion of single-expressing cells into double-expressing cells would not occur ., Instead , co-existence of the three phenotypes in comparable fractions might be observed over a wide range of TGF-β ( Figure 4B ) ., We next considered an asymmetrical model in which the network topology and parameter values differ from the symmetrical model ., In the model with perfect symmetry , we assumed that the inhibitions between Foxp3 and RORγt are equally strong , which is not supported by existing experimental evidence ., In fact , Foxp3 is better known for its inhibitory function on IL-17 , a downstream effector of RORγt , as demonstrated by Williams and Rudensky 29 ., Therefore , we revised our model by removing the direct inhibition of RORγt expression by Foxp3 and adding the inhibition of IL-17 expression by Foxp3 ., This revised model , with broken symmetry ( Figure 1C , Table 1-last column , and Figure 3C ) shows some new features ., First , RORγt behaves ultrasensitively in response to varying TGF-β because of RORγts positive ( autoregulatory ) feedback loop ., Secondly , Foxp3 exhibits multiple saddle-node bifurcations derived from the broken symmetries of the pitchforks ., Interestingly , the four types of stable steady states observed with the symmetrical model have been retained for Foxp3 , and thus for the entire system ., In fact , by varying TGF-β , it is possible to obtain all three differentiated phenotypes in significant fractions simultaneously ., Doing the same analysis for the effect of TGF-β on the induced cell population ( Figure 4C lower panel ) , we found that the asymmetrical model behaved similarly to the symmetrical model ., At low TGF-β , Foxp3 single-positive cells are predicted to be the dominant cell type ., As TGF-β increases to intermediate or high levels , the RORγt single-positive cells and the double-positive cells should appear and co-exist ., These simulation results are in agreement with recently published experimental data documenting the differential effects of TGF-β on the differentiation of TH17 and iTreg cells 16 ., Indeed , at certain intermediate concentrations of TGF-β , three phenotypes in comparable fractions have been observed 16 ., In addition , the maximum percentage of Foxp3 single-positive cells was observed at some lower concentration of TGF-β ., As TGF-β was increased , the percentage of Foxp3 single-positive cells decreased , accompanied by a concordant rise in the percentage of RORγt-expressing cells 16 ., At higher concentrations of TGF-β , RORγt-only cells and double-expressing cells were found to coexist in comparable percentages 15 ., Our model not only validates existing published data on the coexistence of two or more phenotypes in mixed T helper cell populations but also predicts that increasing TGF-β concentration will cause the transformation of Foxp3 single-positive cells into RORγt-expressing cells ., Conversely , decreasing TGF-β concentration might result in the reverse transformation ., We next simulated the influence of IL-6 on this reciprocal differentiation system ., In the asymmetrical model ( Figure 3C ) , IL-6 activates STAT3 , which favors production of RORγt over Foxp3 ., In this model , IL-6 will not trigger differentiation in the absence of TGF-β ., However , IL-6 significantly increases the fraction of RORγt-only cells over a wide range of TGF-β concentrations ( Figure 4A ) ., Also , it stimulates some of the cells in the ( simulated ) population to produce IL-17 ., These results are consistent with the observations of a few groups 13 , 16 ., In particular , Zhou et al . observed that low level TGF-β favors the RORγt-only phenotype and IL-17 production , whereas higher concentrations of TGF-β inhibit the production of IL-17 ., They also reported that the decrease of IL-17 production at higher TGF-β concentration is accompanied by an increase of Foxp3-expressing cells ., We see this phenomenon in our simulation , and we further suggest that the decrease of RORγt-only cells , or the increase of the double-expressing cells , accounts for the reduced production of IL-17 at high TGF-β concentration , because double-expressing cells are known to be much less effective in producing IL-17 than the RORγt-only cells , at least in this type of in vitro assay with TGF-β and IL-6 15 , 16 ., However , Zhou et al . observed a pronounced inhibition of IL-17 production at higher TGF-β concentration even when Foxp3 expression had not been remarkably raised 16 ., This discrepancy suggests that high TGF-β level may trigger Foxp3-independent repression of IL-17 production ., Both the observations by Zhou et al . and our simulations demonstrate that only a minor fraction of RORγt-only cells exhibit IL-17 production even in the presence of IL-6 ., In fact , this is not an idiosyncratic phenomenon ., Mariani et al . recently discovered that only a subset of TH2 cells produce IL-4 due to cell-to-cell variability 30 , suggesting that the production of lineage-specific cytokines in T helper cells can be controlled by stochastic mechanisms ., In the asymmetrical model ( Figure 3C ) , ATRA favors production of Foxp3 over RORγt ., Hence , in our simulation of TGF-β+ATRA stimulation , we found that the percentage of Foxp3-only cells and double-expressing cells significantly increased as compared to TGF-β alone ( compare Figure 4B to Figure 3C ) ., Like IL-6 , ATRA did not trigger differentiation by itself ., We next checked if ATRA can suppress the polarizing effect of IL-6 ., In our simulation , ATRA was effective in reducing the IL-6 induced production of IL-17 ., In addition , at high TGF-β concentration , ATRA significantly decreased the percentage of RORγt-only cells , and resulted in a population with comparable fractions of RORγt-only cells and double-expressing cells ( Figure 5C ) ., All of these simulation results are consistent with published data 13 , 15 , 17 , 31 ., Our model suggests that ATRA can significantly increase the percentage of Foxp3-only cells at intermediate TGF-β concentration , and the percentage of double-expressing cells at high TGF-β concentration ., With our model , we next checked whether IL-6 could reprogram differentiated iTreg cells into TH17 cells ., We first induced a population of naïve CD4+ T cells to differentiate into a population dominated by ‘Foxp3-only’ cells with an intermediate level of TGF-β ( 0 . 28 units ) ., After the cells came to their Foxp3-only steady state , we raised the IL-6 signal to 10 units and continued the simulation ., We found that almost all the cells expressing Foxp3 before adding IL-6 stopped producing Foxp3 upon the treatment with IL-6 , and a subset of ‘RORγt-only’ cells dominated the population ., A fraction of these RORγt-only cells produced IL-17 ( Figure 6A , left panel ) ., When we induced the differentiation of iTreg cells with TGF-β+ATRA and performed the same reprogramming simulation , we found that ATRA did not prevent the repression of Foxp3 expression by IL-6 significantly ., However , ATRA prevented the formation of IL-17 producing cells ( Figure 6A , right panel ) ., The reprogramming capability of IL-6 and the inhibitory effect of ATRA have been observed by Yang et al . 13 ., Analyzing the concentration dependence of these reprogramming effects , we found that a high level of IL-6 may exclusively down-regulate Foxp3 expression ( Figure 6B , left panel ) whereas a high level of ATRA may predominantly prevent IL-17 expression ( Figure 6B , right panel ) ., Interestingly , when both of these factors are present in high concentration , our model predicts that , although most cells exhibit high expression of RORγt , there are almost no IL-17-producing cells in the population ., Future experimental studies are warranted to confirm these intriguing predictions ., Table 2 summarizes the observations that are in agreement with our simulation results and the testable predictions that we have made based on the bifurcation analyses and signal-response curves ., Previous mathematical models have shown how differentiation signals can trigger a robust switch during the development of TH1 or TH2 cells 18 , 19 , 20 , 21 , 22 , 23 , 24 ., In particular , earlier modeling studies by Höfer et al . demonstrated how the interactions among transcription factors can create a memory for TH2 lineage commitment and govern the choice of TH1 and TH2 lineages 18 , 19 ., These studies focused on the dynamics of transcription factors within a single ( average ) cell , but the authors also pointed out that cell-to-cell variability in a CD4+ T cell population can be modeled mathematically by introducing parametric variations to the ordinary differential equations ( ODEs ) ., In addition to modeling molecular interactions , the study by Yates et al . related the dynamics of transcription factors to the phenotypic composition of TH1 and TH2 cell populations 20 ., The authors built comprehensive ODE-based models which take into account cell proliferation , intercellular communication , and cell-to-cell variability ., Yates et al . modeled cell-to-cell variability by variations in initial conditions , but we consider parametric variations to be a more important source of cell-to-cell variability ( see Methods ) ., The reciprocal differentiation of TH17 and iTreg cells , although a relatively new research field , has already been shown to exhibit many interesting and unique features , and yet it has not been studied in quantitative detail using mathematical models ., The work presented here reveals some of the intriguing regulatory mechanisms of this differentiation system ., We showed that the four phenotypes of cells , corresponding to four different steady states of the dynamical system , are derived from a pitchfork bifurcation with certain degree of broken symmetry ., A single primary differentiation signal , TGF-β , can give rise to multiple cell types with distinct functions , while other polarizing differentiation signals , such as IL-6 as ATRA , skew the system to particular type ( s ) of cells ., If we regard TGF-β as tossing dice for the naïve cells , those polarizing signals may load the dice , although they may not toss the dice themselves ., The remarkable advantage of this system is that functionally synergic cells could be generated simultaneously in desired fractions with some simple differentiation inducers ., Our model suggests that the double-expressing phenotype is a re-stabilized co-expressing state , which should be observed in relatively late stages of cell differentiation ., Previously , van den Ham and de Boer found this type of state in a similar dynamical system , although they chose parameter values to avoid this state for their system 24 ., With perfectly symmetrical models , some other groups described a double-expressing state as an intermediate state before the decision making switch , corresponding to some bipotent precursor cells 27 , 32 , 33 ., For the TH17-iTreg paradigm , it is also possible that these double-expressing cells are at an intermediate state that should be converted into single-expressing cells at a later stage of the differentiation process ., However , we do not favor this view for the following reasons ., 1 ) A few studies have shown that the double-expressing cells are effective in repressing effector cell growth and/or secreting pro-inflammatory and anti-inflammatory cytokines 15 , 34 ., It is not likely that a differentiation intermediate would perform any conspicuous function in the immune system ., 2 ) There are a few reports demonstrating the conversion from iTreg cells to double-expressing cells 13 , 14 , or from RORγt-only cells to double-expressing cells 15 , and to our knowledge it is not yet established that observable double-expressing cells can be converted into single-expressing cells ., Assuming that differentiation from early stage to late stage is more readily to be observed than the ‘dedifferentiation’ process , these results indicate that the double-expressing cells might be at a differentiation stage later than the single-expressing states ., 3 ) As shown in this report , there is a mathematical basis to support the double-expressing state appearing only at relatively high TGF-β concentration and some late differentiation stage , and the model is in accord with most published experimental observations ., In addition , we are aware that the double-expressing cells are also observed for iTreg-TH1 and iTreg-TH2 paradigms 3 ., Therefore , the framework presented here may be helpful for understanding iTreg cells that express T-bet or GATA3 as well ., Interestingly , conversion of Foxp3-expressing iTreg cells to Foxp3/T-bet double-expressing cells has been reported 35 ., In fact , these double-expressing cells may play very specific and indispensable roles in controlling inflammation ., Chaudhry et al . have found that iTreg cells require STAT3 for their suppressive function on TH17 , and not on other lineages 36 ., Koch et al . discovered that the T-bet expression is required for the function of iTreg cells during TH1-mediated inflammation 35 ., These results suggest that there are subpopulations of iTreg cells expressing various master regulators of T helper cells , and they are tailored for different functions 3 ., Therefore , the double-expressing cells might be terminally differentiated effectors performing specific suppressive functions ., It is possible that the Foxp3-only cells , which mainly appear at low TGF-β concentration , could serve as precursors or reservoir for different terminal effectors , in addition to their general suppressive functions ., Although the detailed physiological significance of this delicate differentiation system is yet to be discovered , Lochner et al . have already demonstrated in mice that , during infections and inflammation , the number of IL-17 producing RORγt+ cells and double-expressing cells increased in remarkably comparable proportions 15 ., This suggests the need for balance between different cell types in response to pathogenic challenges ., A single differentiation network that gives rise to multiple phenotypes might be crucial for the maintenance of such balance ., Furthermore , it is worth highlighting the common features shared by the TH17-iTreg differentiation system and the differentiation control systems of hematopoietic cells and of stem cells 27 , 28 , 37 ., Functionally , these systems have the potential to generate multiple phenotypes in a single differentiation event , and these phenotypes may play synergic roles under certain physiological conditions ., In addition , it has been shown that cell-to-cell variability within clonal populations makes significant contributions to the stochasticity of lineage choice in stem cells 38 ., This is also concordant with our basic assumptions ., Pitchfork bifurcations ( with broken symmetry ) may be the underlying mechanism generating variable phenotypes in these dynamical control systems ., We will not be surprised if other cell differentiation systems possess similar properties ., Recently , Heinz et al discovered that the ‘priming factor’ PU . 1 , which is required for both macrophage and B cell differentiation , is responsible for creating some of the lineage specific epigenetic markers by itself 39 ., Therefore , it is possible that these priming factors not only drive the differentiation event , but also help to create a heterogeneous population of cells ., One limitation of our model is the assumption that the high concentration of TGF-β used by Lochner et al . is above the saturation concentration for TGF-β signaling 15 ., We are cautious about extrapolating our model to even higher TGF-β concentration because there is no available experimental result for us to compare with ., In fact , it is possible that at even higher TGF-β concentration either the RORγt-only phenotype or the double-expressing phenotype dominates the population , and the conversion between these two phenotypes might be possible by adjusting the concentration of TGF-β ., Although Lochner et al . observed the conversion of RORγt-only cells into double-expressing cells at late time points of induced differentiation , we are not sure about the nature of this conversion: it could be a transition from a transient intermediate to a stable steady state; it could be a transition triggered by a slow increase of TGF-β signaling in RORγt cells , possibly mediated by paracrine signaling ( see below ) ; or it may be caused by slow fluctuations in the transcriptomes 38 ., Nonetheless , when more experimental results become available , we should be able to pinpoint the missing pieces in this reciprocal differentiation system and make the mathematical model more helpful for our understanding of the system in detail ., Another limitation of this study is that we have neglected the effects of intercellular communication on the differentiation of CD4+ T cells ., Cytokines secreted by TH1 and TH2 cells are known to influence the differentiation of neighboring T cells 40 , and previous modeling work has highlighted the importance of these paracrine signaling effects 20 ., Relevant to our work , the cytokines secreted by TH17 and iTreg cells can influence the differentiation of a population of T cells , and this influence might be reflected in changes of the proportions of induced phenotypes ., For example , both TH17 and iTreg cells can produce TGF-β 41 , 42 , which may increase the percentage of both type of cells , or induce the transition from single-expressing cells to double-expressing cells , and this may be causative for the transition observed by Lochner et al . 15 ., However , it is not yet clear how important are paracrine signals via secreted cytokines compared to exogenous cytokine signals , with respect to TH17 and iTreg differentiation ., We leave the consideration of these factors for future work ., In summary , we presented a novel mathematical model of TH17-iTreg differentiation ., Based on the model , we show how TGF-β can trigger the differentiation of naïve CD4+ T cells into a heterogeneous population containing RORγt-only , Foxp3-only and double-expressing cells , and how polarizing signals can skew the differentiation to particular phenotype ( s ) ., The model suggests how the conversions among different phenotypes can be guided ., Additionally , the model gives a new quantitative explanation for the double-expressing cells , which should appear only at a late stage of the differentiation process ., Our model provides new insights into the regulatory mechanisms that underlie the molecular control of certain immune responses ., We constructed our mathematical model based on known interactions among key molecules in the differentiation system of TH17 and iTreg cells ., For illustrative purposes , we first consider a ‘symmetrical’ model in which the lineages of TH17 and iTreg have identical corresponding interaction types and strengths ., Then we added two intermediate proteins for transmitting TGF-β signals in this symmetrical model ., Next , we modified our model so that it became asymmetrical , and we incorporated two other input signals ., Using this last model , we compared our simulation results with some published experimental data and made several testable predictions ., In the symmetrical model ( Figure 2A ) TGF-β upregulates both RORγt and Foxp3 , which has been demonstrated in a few published experiments 13 , 43 ., The model includes the ‘autoactivation’ of both master regulators ., Although there is no evidence for direct autoactivation of RORγt and Foxp3 , these relationships in our model represent known positive feedback loops in their respective pathways ., One origin of these positive feedback loops is

Introduction, Results, Discussion, Methods

The reciprocal differentiation of T helper 17 ( TH17 ) cells and induced regulatory T ( iTreg ) cells plays a critical role in both the pathogenesis and resolution of diverse human inflammatory diseases ., Although initial studies suggested a stable commitment to either the TH17 or the iTreg lineage , recent results reveal remarkable plasticity and heterogeneity , reflected in the capacity of differentiated effectors cells to be reprogrammed among TH17 and iTreg lineages and the intriguing phenomenon that a group of naïve precursor CD4+ T cells can be programmed into phenotypically diverse populations by the same differentiation signal , transforming growth factor beta ., To reconcile these observations , we have built a mathematical model of TH17/iTreg differentiation that exhibits four different stable steady states , governed by pitchfork bifurcations with certain degrees of broken symmetry ., According to the model , a group of precursor cells with some small cell-to-cell variability can differentiate into phenotypically distinct subsets of cells , which exhibit distinct levels of the master transcription-factor regulators for the two T cell lineages ., A dynamical control system with these properties is flexible enough to be steered down alternative pathways by polarizing signals , such as interleukin-6 and retinoic acid and it may be used by the immune system to generate functionally distinct effector cells in desired fractions in response to a range of differentiation signals ., Additionally , the model suggests a quantitative explanation for the phenotype with high expression levels of both master regulators ., This phenotype corresponds to a re-stabilized co-expressing state , appearing at a late stage of differentiation , rather than a bipotent precursor state observed under some other circumstances ., Our simulations reconcile most published experimental observations and predict novel differentiation states as well as transitions among different phenotypes that have not yet been observed experimentally .

In order to perform complex functions upon pathogenic challenges , the immune system needs to efficiently deploy a repertoire of specialized cells by inducing the differentiation of precursor cells into effector cells ., In a critical process of the adaptive immune system , one common type of precursor cell can give rise to both T helper 17 cells and regulatory T cells , which have distinct phenotypes and functions ., Recent discoveries have revealed a certain heterogeneity in this reciprocal differentiation system ., In particular , treating precursor cells with a single differentiation signal can result in a remarkably diverse population ., An understanding of such variable responses is limited by a lack of quantitative models ., Our mathematical model of this cell differentiation system reveals how the control system generates phenotypic diversity and how its final state can be regulated by various signals ., The model suggests a new quantitative explanation for the scenario in which the master regulators of two different T cell lineages can be highly expressed in a single cell ., The model provides a new framework for understanding the dynamic properties of this type of regulatory network and the mechanisms that help to maintain a balance of effector cells during the inflammatory response to infection .

systems biology, theoretical biology, immunology, biology, computational biology, molecular cell biology, genetics and genomics

journal.pntd.0006967

Detection of clinical and neurological signs in apparently asymptomatic HTLV-1 infected carriers: Association with high proviral load

HTLV-1 , a human retrovirus , is the causative agent of Adult T Leukemia/Lymphoma ( ATLL ) and HTLV-1-associated myelopathy ( HAM/TSP ) 1 , at least 5–10 million people infected worldwide , almost 5–10% of them in Brazil 2 ., However high , such numbers may be an underestimate since only 2/3 of the world has been mapped for HTLV infection 3 ., Clinically , HAM/TSP is characterized by muscle weakness , hyperreflexia , spasticity in the lower extremities and urinary disturbances associated with preferential damage to the thoracic spinal cord 4 ., HTLV-1 has been shown to be associated not only with HAM/TSP but also with several inflammatory diseases , such as alveolitis , polymyositis , arthritis , infective dermatitis , Sjögren syndrome and uveitis 5–9 ., In addition , sensory and gait abnormalities , isolated bladder dysfunction , erectile dysfunction , and sicca syndrome , have all been reported among HTLV-1–infected individuals without HAM/TSP 10 ., Several neurological manifestations that are not explained by myelopathy have been described in so-called symptomatic persons , such as peripheral polyneuropathy , myositis , dysautonomia and cognitive alterations , as well as cranial neuropathies , movement disorders and an amyotrophic lateral sclerosis ( ALS ) -like syndrome 11 ., Despite the fact that few patients ( <10% ) will develop classical syndromes ( ATLL and HAM/TSP ) , preliminary observations indicate that other symptoms and subclinical neurological disturbances can develop in those individuals 12 , but they have not been well-defined as new clinical outcomes related to early inflammation process ., In addition , peripheral neuropathy is significantly more frequent in the seropositive group ., In a study with 153 HTLV-1-infected carriers , the presence of higher frequency of motor and bladder dysfunctions in HTLV-1 patients as compared with uninfected control subjects was found 13 ., Those data suggest that HTLV-1-infected individuals may exhibit a wide variety of neurological manifestations distinct from the classical picture of HAM/TSP 13 ., It is unclear whether such manifestations share a common characteristic with the spinal cord disease ., This study aims to demonstrate that some clinical conditions , neurological finds and HTLV-1 proviral load may be associated with further development of full-blown HAM/TSP , in individuals considered free of the disease according to currently used criteria for its diagnosis ., To do so we studied patients from a large cohort of asymptomatic HTLV-1 carriers who have been followed for more than twenty years ., Clinical evaluation and a standardized screening neurological examination were performed by MH ( a board-certified neurologist , and blinded for HTLV-1 clinical condition ) for all subjects ., Only symptoms/signals already associated with HTLV-I infection in previous reports were considered , and they should have no other clinical explanation ., Each patient had at least one neurological/clinical evaluation , and a standardized questionnaire was used , with separate questions for clinical and neurological aspects 15 ., HAM/TSP diagnostic criteria was based on recommendations from an international consortium 16 ., Briefly , definite HAM/TSP is a non-remitting progressive spastic paraparesis with sufficiently impaired gait to be perceived by the patient ., Sensory symptoms or signs may or may not be present but when present , they are subtle and without a clear-cut sensory level ., Urinary and anal sphincter signs or symptoms may or may not be present , and the presence of anti-HTLV-1 antibodies in the cerebro spinal cord fluid ( CSF ) ., Probable HAM/TSP was defined by a monosymptomatic presentation: spasticity or hyperreflexia in the lower limbs or isolated Babinski sign with or without subtle sensory signs or symptoms , or neurogenic bladder only confirmed by urodynamic tests ., For both definite and probable definitions , clinicians must exclude an array of disorders that can mimic HAM/TSP ., We described a possible intermediate state of HAM not fulfilling the classical definition ( 16 ) ., To be considered as an intermediate syndrome case the patient must present more three signs , found during a neurological evaluation , with the investigator blinded to the patient’s HTLV status ., For this present study , only clinical findings previously associated with HTLV-1 were considered , such as dermatological , ophthalmological , rheumatological , urinary , disautonomic , and oral changes 17 ., Neurological evaluation included tests of strength in upper and lower limbs , cranial nerves function and patellar , biceps and plantar reflexes , as well as an appraisal of the vibration sense ., The presence of minimal changes in muscular strength or gait was explored: subjects were asked to walk on their heels , toes , tandem gait , and rise from a chair without help from their arms ., A large clinical and laboratory database has been organized on an internet based platform using REDCap , software developed at the Vanderbilt University by an informatics core ., All clinical data , which have been updated on a regular basis over the last 20 years , were entered into a specific REDCap database 18 ., HTLV-1 proviral load was quantified by real-time PCR , using primers and probes targeting the pol gene: SK110 and SK111 , the internal HTLV-1 Taq Man probe was selected using Oligo ( National Biosciences ) ., All samples were run in duplicate , and results expressed as HTLV-1 DNA copies/104 peripheral blood mononuclear cells ( PBMCs ) , as described elsewhere 19 ., The Ethical Board of the IIER approved the protocol ( Number 86379218 . 6 . 1001 . 0061 ) ., We obtained signed informed consent from all participants prior to study inclusion , and all participants were adults ., Statistical analysis was conducted using Student’s t-test for parametric data , and the chi-square test for proportions ., Bivariate logistic analysis was performed to identify independent variables associated with the intermediate syndrome ( IS ) ., Variables associated with the outcome at a significance level of p<0 . 20 ( IS ) in the bivariate analysis were included in a multivariate logistic model , in a stepwise forward fashion ., Such variables were: visual symptoms , skin lesions , oral conditions , bladder disfunction , and rheumatological conditions ., The best fitting model was selected ., The logistic analysis was performed with the aid of Stata 12 software ( StataCorp . 2011 . Stata: Release12 . Statistical Software . College Station , TX ) ., We enrolled 175 HTLV-1 patients on this study and classified them as having or not criteria for the diagnosis of the intermediate syndrome ., Based on a thorough neurological examination , 42 patients met the criteria for making the diagnosis of the intermediate syndrome , whereas 133 did not and were called “asymptomatic” ( not having the intermediate syndrome ) ., All of them had intermediate symptoms that were classified as probable HAM/TSP at entry , primarily neurogenic bladder confirmed by urodynamic study ., Table 1 shows the univariate analyses of socio demographic variables and proviral load of all volunteers; mean age of the enrolled subjects ( n = 175 ) was 46 . 3 years and 130 ( 74 . 3% ) were females ., Most of the patients were white ( 56 . 5% ) , and the PVL from the IS cases was six times that from patients without IS ( p<0 . 001 ) ., Clinical classificafion on Table 2 shows that neurologic symptoms/signals ( p<0 . 001 ) , visual disorders ( p = 0 . 001 ) , oral manifestations ( p = 0 . 001 ) , skin lesions ( p<0 . 001 ) , bladder disorders ( p<0 . 001 ) , and rheumatologic symptoms ( p = 0 . 001 ) , were strongly associated to IS , except for disautonomy ( p = 0 . 21 ) ., On a multivariate model analysis , including gender , age , and PVL and several clinical conditions , such as oral conditions , bladder disorders and rheumatological symptoms were independently associated with SI outcome ., In this same model , all these variables and age were also significantly associated with the outcome , when included as continuous variables ( Table 3 ) ., Table 4 shows that the presence of more than three or more signs and/or symptoms was significantly associated with the intermediate syndrome ( p = 0 . 006 ) , therefore the the cut-off point for that diagnosis was set at this point ., This study aimed to define the early neurological disorders that can be present in HTLV-1-infected subjects ., We found 24% of HTLV-1-infected patients from our outpatient service who were initially considered asymptomatic to have enough signs and symptoms putting them on a novel category , called intermediate syndrome ., The correlation between some of their symptoms and the proviral load also reinforces the importance of such mild forms , which may constitute either an independent clinical intermediate syndrome or markers for an early diagnosis of HAM/TSP ., A significantly PVL was also present in patients presenting three or more symptoms or signs ., HAM/TSP is a chronic progressive myelopathy characterized by bilateral pyramidal tract involvement with sphincter disturbances ., Why only a small proportion of HTLV-1-infected individuals develops classical HAM/TSP is not known 11 ., The main neurological symptoms of the disease are progressive and lead to deterioration in the quality of life , but minor neurological symptoms can also be found among HTLV-1 carriers 11 ., In a study with 153 HTLV-1-infected carriers and 388 HTLV-2-infected subjects , the presence of neurological abnormalities was prospectively ascertained , with a higher frequency of motor and bladder dysfunctions in HTLV-1 as compared with uninfected control subjects 10 ., All those data suggest that HTLV-1-infected individuals can exhibit a wide variety of neurological manifestations distinct from the classical picture of HAM/TSP 11 ., It is unclear whether those manifestations share a common characteristic with this diagnosis ., We and others have shown a correlation between proviral load and Tax gene expression with the presence of HAM/TSP 14 , 19 , 20 ., We hypothesize that the burden of HTLV-1 was correlated with neurological disturbances that fall short of HAM/TSP , and with cognitive dysfunction ., Demonstration of that hypothesis might provide a link between HTLV-1 burden and early neurological dysfunctions , providing impetus to the development of methods aiming to reduce the proviral load in infected subjects ., Perhaps the explanation for the observed neurological findings lies in the white matter ., The white matter present in the CNS has the important function of transporting neural signals from subcortical regions to the cortex and from the cortex to the subcortical regions ., In the CNS , ischemic and traumatic injuries often result in significant functional deficit , most of which can be attributed to white matter dysfunction 21–23 ., In the case of HTLV-1 , there is participation of components of the inflammatory response on the mechanisms underlying the demyelination process characteristic of this disease 20 ., We found that HTLV-1 infection is associated with a variety of clinical manifestations occurring in patients who either do not have or who did not have developed full HAM/TSP yet ., The correlation between some of their symptoms and the proviral load also reinforces the importance of such milder forms , which may constitute either an independent clinical SI or markers for an early diagnosis of HAM/TSP ., A significantly higher proviral load was present in patients presenting more than three symptoms/signs , a cut-off point that can constitute a surrogate marker for clinical progression ., In conclusion , this preliminary report identified the presence of some clinical and neurological symptoms , in subjects classified originally as asymptomatic , which may be promising markers for early HAM/TSP progression ., This knowledge may contribute for a stricter clinical vigilance of the so-called asymptomatic HTLV-1 positive patients , prompting the introduction of a treatment for the infection , if and when it becomes available in the future .

Introduction, Methods, Results, Discussion

Several studies suggest that HTLV-1 infection may be associated with a wider spectrum of neurologic manifestations that do not meet diagnostic criteria for HAM/TSP ., These conditions may later progress to HAM/TSP or constitute an intermediate clinical form , between asymptomatic HTLV-1 carriers and those with full myelopathy ., Our aim was to determine the prevalence of HTLV-1-associated disease in subjects without HAM/TSP , and the relationship between these findings with HTLV-1 proviral load ( PVL ) ., Methods: 175 HTLV-1-infected subjects were submitted to a careful neurological evaluation , during their regular follow up at the HTLV outpatient clinic of the Institute of Infectious Diseases “Emilio Ribas” , São Paulo city , Brazil ., Clinical evaluation and blinded standardized neurological screening were performed for all the subjects by the same neurologist ( MH ) ., Results: After the neurological evaluation , 133 patients were classified as asymptomatic and 42 fulfilled the criteria for intermediate syndrome ( IS ) ., The mean age of the enrolled subjects was 46 . 3 years and 130 ( 74 . 3% ) were females ., Clinical classification shows that neurological symptoms ( p<0 . 001 ) , visual disorders ( p = 0 . 001 ) , oral conditions ( p = 0 . 001 ) , skin lesions ( p<0 . 001 ) , bladder disorders ( p<0 . 001 ) , and rheumatological symptoms ( p = 0 . 001 ) , were strongly associated to IS , except for disautonomy ( p = 0 . 21 ) ., A multivariate analysis revealed that HTLV-1 proviral load , oral conditions , bladder disorders and rheumatological symptoms were independently associated with the IS ., Conclusions: We found some early alterations in 42 patients ( 24% ) , particularly the presence of previously not acknowledged clinical and neurological symptoms , among subjects previously classified as asymptomatic , who we reclassified as having an intermediate syndrome .

At least 5–10 million people live with the Human T-Cell Lymphotropic Virus type 1 ( HTLV-1 ) worldwide , and around 0 . 25–5% of them may develop HTLV-1-associated myelopathy/Tropical spastic paraparesis ( HAM/TSP ) , which is associated with chronic inflammation ., In this study , involving 175 HTLV-1-infected subjects originally classified as asymptomatic , we found that 42 of them in reality presented some early clinical conditions , including alterations related not only to the neurological system , but also to the eyes and the skin ., We called such conditions intermediate syndrome ., Thus , it seems reasonable to suggest that all HTLV-1-infected subjects should be monitored for symptoms that may arise earlier in the course of their infection .

urology, cognitive neurology, medicine and health sciences, bladder and ureteric disorders, pathology and laboratory medicine, nervous system, neurodegenerative diseases, pathogens, amyotrophic lateral sclerosis, bladder, microbiology, neuroscience, retroviruses, viruses, cognitive neuroscience, rna viruses, signs and symptoms, neurological signaling, motor neuron diseases, medical microbiology, htlv-1, microbial pathogens, lesions, signal transduction, diagnostic medicine, anatomy, cell biology, central nervous system, viral pathogens, neurology, biology and life sciences, renal system, cognitive science, cell signaling, organisms

journal.pcbi.1004187

A Bayesian Ensemble Approach for Epidemiological Projections

Epidemiological forecasting is inherently challenging because the outcome often depends on largely unpredictable characteristics of hosts and pathogens as well as contact structure and pathways that mediate transmission 1 ., Faced with such uncertainty , policy makers must still make decisions with high stakes , both in terms of health and economics ., For instance , global annual malaria mortality was recently estimated at around 1 . 1 million 2 and to optimize control efforts , policy makers must make seasonal predictions about spatiotemporal patterns 3 ., The prospect of an emergent pandemic influenza outbreak remains a global threat and emergency preparedness must evaluate the costs and benefits of control measures such as border control , closing of workplaces and/or schools as well as different vaccination strategies 4 ., Livestock diseases are major concerns for both animal welfare and economics ., As an example , the United Kingdom ( UK ) 2001 outbreak of foot and mouth disease ( FMD ) involved culling of approximately 7 million animals , either in an effort to control the disease or for welfare reasons , and the total cost has been estimated at £8 billion 5 ., To minimize the size and duration of future outbreaks , various strategies for culling and vaccination must be compared 6–8 ., As a tool to address these challenging tasks , mathematical models offer the possibility to explore different scenarios , thereby informing emergency preparedness and response to epidemics 1 , 9–12 ., The predictive focus of epidemiological models can either be classified as forecasting or projecting 13 ., Forecasting aims at estimating what will happen and can be used for example to predict seasonal peaks of outbreaks 3 , 14 or to identify geographical areas of particular concern 15 ., Projecting , which is the main focus of this study , instead aims at comparing different scenarios and exploring what would happen under various assumptions of transmission , e . g . comparing the effectiveness of different control actions 7 , 16–19 ., Whilst analytical models clearly provide important insight into observed dynamics and a theoretical understanding of epidemiology 20–22 , there has been a shift in recent years towards stochastic simulation models for predictive purposes 1 ., Typically , dynamic models are constructed and outbreaks are simulated repeatedly , thus generating predictive distributions of outcomes 1 , 17 , 18 , 23 ., This variability in outcomes caused by the mere stochasticity of the transmission process includes one level of uncertainty , but still only relies on a single set of assumptions about the underlying disease transmission process ., However , multiple assumptions can often be justified , leading to further uncertainty in the predictions ., For instance , different models may have different projections because of different assumptions about transmission or because they incorporate different levels of detail ., It may also be informative to explore different projections in terms of different parameterizations of a single model , for example corresponding to worst or best case scenarios ., Faced with a set of projections , an important issue is how to combine these in a manner such that they can be used to inform policy ., The issue of multiple projections is not unique to the field of epidemiology , and various techniques of ensemble modeling have been used to merge projections based on different modeling assumptions ., The key concept is that rather than relying on a single set of assumptions , a range of projections is used for predictive purposes ., For instance , climate forecasting has employed ensemble techniques to account for uncertainty about initial conditions , parameter values and structure of the model design when predicting climate change 24 , 25 ., Weather forecasting has been improved by combining the results of multiple models 26 , 27 ., Similarly , hydrological model ensembles have been demonstrated to increase reliability of catchment forecasting 28 and have been used to assess the risk of flooding events 29 ., Ensemble methods have also proven to be a powerful decision tool for medical diagnostics 30 , 31 and ecological considerations including management 32 and prediction of future species distribution 33 ., Ensemble modeling has not yet been extensively used in epidemiology ., However a few implementations exist , commonly by feeding climate or weather ensembles into disease models ., Daszak et al . 34 coupled a set of climate projections to an environmental niche model of Nipah virus to predict future range distribution of the virus under climate change ., Similarly , Guis et al . 35 investigated the effect of climate change on Bluetongue emergence in Europe by simulating outbreaks under different climate change scenarios ., Focusing on a shorter time scale , Thomson et al . 3 used an ensemble of seasonal forecasts to predict the spatiotemporal pattern of within seasonal variation in malaria incidence ., These studies all used a single disease model projection , coupled to an ensemble of climate or weather forecasts and the use of structurally different epidemiological models are to our knowledge still rare ., However , Smith et al . 36 compared different malaria vaccination strategies by implementing an ensemble approach with different alterations of a base model ., Also , in order to estimate global malaria mortality , Murray et al . 2 used weighted averages of different predictive models ., Given the success of ensemble methods in other fields , we expect that epidemiological implementations will increase ., For that purpose however , there is a need for methods that combine multiple projections ., A central issue in ensemble modeling is how to weight different projections , and we envisage four main procedures for this ., Firstly , all models can be given equal weights ., For instance , the IPCC 2001 report on climate change 37 used a set of climate models and gave the range of probable scenarios by averaging over different models and uncertainty by envelopes that included all scenarios ., Gårdmark et al . 32 used seven ecological models for cod stock and argued that in order to prevent underestimation of uncertainty , weighted model averages are not to be used and when communicating with policy makers , it is preferable to present all included projections as well as the underlying assumptions behind them ., A similar approach was also used by Smith et al . 36 , who presented the prevalence of malaria under different vaccination strategies by medians of individual models and the range of the whole ensemble ., Secondly , expert opinions can be used to weight models ., To our knowledge , no ensemble study has implemented weights based exclusively on expert opinion , but Bayesian model averaging can incorporate expert opinion as a subjective prior on model probabilities 38 ., This approach relies on engaging stakeholders and communicating the underlying assumptions of the projections ., Thirdly , models can be weighted by agreement with other models ., This approach was implemented by Räisänen and Palmer 39 , who used cross-validation to weight climate models ., As a more informal approach to the use of model consensus , the third IPCC report excluded two models because these predicted much higher global warming than the rest of the ensemble , thus acting as outliers 24 ., Fourthly , weights can be determined by the models’ ability to replicate data ., If all models are fitted to the same data using likelihood based methods , weights can be given directly by Akaike or Bayesian Information Criterion ( AIC or BIC ) 40 , 41 ., In the FMD context , this may be a suitable approach if all included models are data driven kernel models that estimate parameters from outbreak data , such as those proposed by Jewell et al . 42 or Tildesley et al . 43 ., However , such weighting schemes would be unfeasible when including detailed simulation models that rely on a large number of parameters , that are determined by expert opinion or lab experiment , such as AusSpread 44 , NAADSM 45 and InterSpread Plus 46 ., We propose that the future of ensemble modeling for epidemiology will benefit from combining structurally different model types , and methods of weighting need to handle both kernel type models as well as detailed simulation models ., Thus , bias assessment is often confined to the ability of models to replicate observed summary statistics of interest , in particular when the resolution of data observation is on a courser scale than the model prediction 47 ., Such methods have been developed within the field of climate forecasting ., Giorgi and Mearns 48 introduced a formal framework in which model weights were assessed based on model bias compared to observed data as well as convergence , i . e . agreement with the model consensus ., Tebaldi et al . 47 extended the approach to a Bayesian framework ., This approach is appealing because it provides probability distributions of quantities of interest , hence uncertainty about the projected outcomes may be provided to policy makers ., As such , it would be a suitable approach also for epidemiological predictions ., However , methods developed in one field might not be directly transferable to another ., Tebaldi et al . 47 points out that lack of data at fine scale resolution is a limiting factor for climate forecasting ., Yet , at courser resolution climate researchers have access to long time series of climate variables to assess model bias ., Comparable data may be available for endemic diseases , such as malaria 36 or tuberculosis 49 , or seasonally recurrent outbreaks , such as influenza 14 or measles 50 ., However , for emerging diseases , long time series would rarely be available , making the lack of data an even bigger issue for epidemiology ., In this methodological paper we aim to explore the potential of using ensemble methods based on the approach presented by Tebaldi et al . 47 for epidemiological projections ., The Tebaldi et al . methodology focus on ensembles where projections are made with different models and our aim is to provide a corresponding framework for disease outbreak projections ., To investigate the potential of the framework for epidemiology , we here use variations of a single model as a proxy for different models , thus allowing us to investigate how the methodology performs under different levels of discrepancy among projections in the ensemble ., We exemplify the implementation by using the UK 2001 FMD outbreak and projections modelled by different parameterizations of the Warwick model 7 , 9 ., In the 2001 UK FMD outbreak , livestock on all infected premises ( IPs ) were culled ., In addition , livestock on farms that were identified to be at high risk of infection were culled as either traditional dangerous contacts ( DCs ) or contiguous premises ( CPs ) ., CPs were defined as “a category of dangerous contact where animals may have been exposed to infection on neighboring infected premises” 5 , 8 ., We start by focusing on ensemble prediction of epidemic duration under the control action employed during the 2001 outbreak compared with an alternative action that excludes CP culling ., We investigate sensitivity to priors and explore a hierarchical Bayesian extension of the method to circumvent potential problems with prior sensitivity ., We also explore the potential of including subjective a priori trust in the different modeling assumptions and extend the methodology further to allow incorporation of multiple epidemic quantities , here exemplified by adding number of infected and culled farms to the analysis ., Through a simulation study , we finally explore the capacity and limitations of the proposed ensemble method , pinpointing some important features of ensemble modeling, We focus on projections of FMD made by the Warwick model 7 , 9 ., This model was developed in the early stages of the 2001 FMD outbreak by Keeling and coworkers to determine the potential for disease spread and the impact of intervention strategies 9 ., Here , we utilized a modified version of the model used in 2001 , and we briefly describe relevant aspects of the Warwick model with regard to ensemble modeling ., Full details of the model can be found in 7 , 9 ., The rate at which an infectious farm I infects a susceptible farm J is given by:, RateIJ=SusJ×TransI×K ( dIJ ), ( 1 ), where, SusJ= ( Zsheep , Jps , SSsheep+Zcow , Jpc , SScow ), ( 2 ), is the susceptibility of farm J and, TransI= ( Zsheep , Ips , TTsheep+Zcow , Ipc , TTcow ), ( 3 ), is the transmissibility of farm I and K ( dIJ ) is the distance-dependent transmission kernel , estimated from contact tracing 9 ., In this model Zs , I is the number of livestock species s ( sheep or cow ) recorded as being on farm I , Ss and Ts measure the region and species-specific susceptibility and transmissibility , dIJ is the distance between farms I and J and K ( dIJ ) is the distance dependent transmission kernel ., The parameters , ps , S , pc , S , ps , T and pc , T , are power law parameters that account for a non-linear increase in susceptibility and transmissibility as animal numbers on a farm increase ., Previous work has indicated that a model with power laws provides a closer fit to the 2001 data than when these powers are set to unity 43 , 51 , 52 ., This version of the model has previously been parameterized to fit to the 2001 FMD outbreak 43 ., Region-specific transmissibility and susceptibility parameters ( and associated power laws ) capture specific epidemiological characteristics and policy measures used in the main hot spots of Cumbria , Devon and the Welsh and Scottish borders ., The model is therefore fitted to five regions: Cumbria , Devon , Wales , Scotland and the rest of England ( excluding Cumbria and Devon ) ., A table listing all the parameter values used in this model is given in Tildesley et al . 43 ., In order to obtain multiple modeling assumptions for ensemble modeling , we specified different transmission rates , i as, RateIJi=SusJ×k1iTransI×K ( k2idIJ ), ( 4 ), where k1i and k2i are constants , specific for each modeling assumption , that scale the transmissibility and the spatial kernel , respectively ., k1i = k2i = 1 , follow the parameterizations of Tildesley et al . 43 and by decreasing or increasing these constants , we obtain parameterizations that correspond to best or worst case expectations about the transmissibility and spatial range of transmission ., We are interested in how the level of discrepancy among modeling assumptions influences the performance of the ensemble method and we therefore created two different ensembles with different k1i and k2i , as listed in Table 1 . We refer to these as the large and small discrepancy ensemble , corresponding to large and small differences , respectively , in the underlying modeling assumptions used for projections ., DCs in our model were determined based upon both prior infection by an IP and future risk of infection in the same way as in previous work 8 ., CPs were defined as farms that share a common boundary and were determined on an individual basis ., The model was seeded with the farms that were predicted to be infected prior to the introduction of movement restrictions on the 23rd February ., For each modeling assumption i and control action , 200 replicates were simulated and each simulation progressed until the epidemic died out ., To demonstrate concepts and explore the potential of using the Tebaldi et al . 47 approach for epidemiological considerations we initially focus on outbreak duration ., This is often considered to be the most costly aspect of FMD outbreaks due to its implication for trade 53 ., In section 2 . 7 we extend the methodology to multiple epidemic quantities ., However , the outbreak duration example offers transparent transition from the original climate analysis of Tebaldi et al . 47 that considers the ensemble estimated difference between current and future mean temperatures ., In order to introduce the framework to epidemiology , we consider the difference between the implemented and an alternative control action , attempting to show whether the inclusion of CP culling was an appropriate choice given the conditions at the start of the outbreak ., As this is a post outbreak analysis , we know the final outbreak duration of the observed outbreak , but that is just a single realization and due to the stochastic nature of disease transmission , the exact outcome may be quite variable ., We also have no observed outbreak under the alternative control action to compare with the implemented control ., Under these conditions , the most appropriate quantities to compare are the mean duration of a large number of outbreaks under the two control actions , something that can only be achieved through epidemic modeling ., We are interested in comparing projections under the implemented control action to the observed data in order to estimate model weights ., Using the Bayesian method of Tebaldi et al . 47 , weights as well as statistics of the outbreak , like duration , are considered unknown random variables , and we denote the mean outbreak duration under the implemented and the alternative control action as μ and v , respectively , corresponding to the mean current and future temperature , respectively , for the climate application ., In order to fit with the normal assumptions of the method , we consider log-duration in the analysis ., Weights are expressed through precision , λ = λ1 , λ2 , … , λn , with λi denoting the precision of modeling assumption, i . The projection specific parameter xi indicates the mean of all replicates under the implemented control action ( analog of current climate ) for modeling assumption, i . For the UK 2001 outbreak this included culling of IPs , DCs and CPs ., The corresponding projection for the alternative control action ( analog of future climate ) , that included culling of IPs and DCs is denoted yi ., The relationship between projections and ensemble parameters is expressed as, xi~Normal ( μ , λi-1 ), ( 5 ), yi~Normal ( ν+β ( xi-μ ) , ( θλi ) -1 ) ,, ( 6 ), with Normal ( μ , λi-1 ) denoting the normal distribution with mean μ and variance λi-1 ., Parameter θ is included to allow for difference in overall precision of the modeling assumptions under implemented and alternative control actions ., However , since projections xi and yi are based on simulations , it is fair to assume that modeling assumption i that has a high precision for the observed control action also will do well for the unobserved action ., This is incorporated by the λi term in both Eqs 5 and 6 ., For the same reason , we may expect that a projection of a large xi also corresponds to a large value for yi and thus β is included to allow for correlation between corresponding projections for the two control actions; a projection that e . g . over-predicts duration of the outbreak for the observed control action can be expected to also over-predict the alternative control action ., The analysis of Tebaldi et al . 47 also assessed bias of projections by their ability to reproduce observed current climate by incorporating the relationship between observed current climate x0 , an unobserved true mean climate variable ( μ ) and the precision of natural variability τ0 through, x0~Normal ( μ , τ0-1 ) ., ( 7 ), In climate modeling , it is a fair assumption that τ0 is a known , fixed parameter because it can be assessed through historical data ., That would rarely be the case for the corresponding epidemic considerations , at least for emerging diseases ., Using a single outbreak to evaluate bias , we clearly have no way of assessing variability in outcomes ., We therefore include τ0 as an unknown , random variable that is estimated in the analysis as described in the following section ., To aid the interpretation and transfer from the climate to the epidemiological interpretation , we have included Table 2 that lists the variables used in the analysis ., Our main interest in terms of outcome under the implemented control action is μ rather than x0 ., However , it is clear that in addition to the mean duration of the outbreak , the uncertainty about the process also results in some variability in the outcomes that we need to consider ., The stochastic simulations used to generate projections provide not only a mean simulated outbreak quantity , but also a range of outcomes that projects the variability ., In the absence of repeated outbreaks to evaluate variability of outcomes , an obvious choice would be to use this information to inform the variability τ0 ., Defining the variability τi as the precision of projections under the implemented action for modeling assumptions i = 1 , 2 , … , n we include a hierarchical structure in the analysis so that for i = 0 , 1 , 2 , … , n, τi~Gamma ( aτ , bτ ) ,, ( 8 ), where Gamma ( aτ , bτ ) indicates the gamma distribution with shape parameter aτ and rate bτ both of which are unknown parameters and are estimated in the analysis ., Thus , as it would be cumbersome to elicit a fixed prior for τ0 based on our prior expectations about variability , we instead assume that τ0 comes from some , unknown distribution , and make use of τ1 , τ2 , … , τn to inform what this distribution should be ., Similarly , we need to model the variability of projections under the alternative control action , and denoting this φi we specify The parameters aφ and bφ are conditionally independent from all other parameters in the analysis and can be modelled separately in the Bayesian analysis ., As xi , yi , τi and φi are calculated from a finite number of realization with each modeling assumption and control action , there is some uncertainty related to this ., Tebaldi et al . 47 however points out that while it is certainly possible to construct a Bayesian model that takes this uncertainty into account , the effect is minimal if the number of replicates is large ., With the R = 200 replicates preformed here , the uncertainty of the mean will in practice have very little effect , and we have included xi , yi , τi and φi as fixed observations ., Priors for aτ and bτ were specified as a gamma distribution with shapes Aaτ and Abτ , respectively , and rates Baτ and Bbτ , respectively ., Similarly , the priors for aφ and bφ , were specified as a gamma distribution with shapes Aaφ and Abφ , respectively , and rates Baφ and Bbφ , respectively ., We explored different parameter choices for the hyperpriors and found that the results were insensitive to the choice of prior for a wide range of values ., In the analysis presented , we used Aaτ = Abτ = Baτ = Bbτ = Aaϕ Abϕ = Baϕ = Bbϕ = 0 . 001 ., This corresponds to prior distributions with a mean of one and a variance of 1000 , thus allowing for a wide range of plausible values ., Bayesian analysis requires the specification of prior parameters ., We follow Tebaldi et al . 47 with priors specified as uniform on the real line for μ , ν , and β , and λi~Gamma ( aλ , bλ ) for i = 1 , 2 , … , n and θ~Gamma ( aθ , bθ ) ., We also need to specify hyperpriors for aτ and bτ , and we implement aτ~Gamma ( Aaτ , Baτ ) and bτ~Gamma ( Abτ , Bbτ ) ., Denoting x = x1 , x2 , … , xn and y = y1 , y2 , … , yn , the full posterior distribution under these priors is given by, P ( μ , ν , β , λ , θ , τ0|x0 , x , y , τ1 , τ2 , … ) ∝∏i=1n ( λiaλ−1e−bλλiλiθ1/2exp{ −λi2 ( xi−μ ) 2+θ ( yi−ν−β ( xi−μ ) ) 2 } ) θaθ−1e−bθθτ01/2exp{ τ02 ( x0−μ ) 2 }∏i=0n ( τiaτ−1e−bττi ) aτAaτ−1e−BaτaτbτAbτ−1e−Bbτbτ, ( 10 ), This posterior only differs from the one defined by Tebaldi et al . in that we include τ0 as an unknown variable and use a hierarchical structure for its prior ., Using Markov Chain Monte Carlo ( MCMC ) techniques as described in 2 . 9 , we first performed the analysis with priors as specified by Tebaldi et al . 47 where applicable ,, i . e ., aλ = bλ = aθ = bθ = 0 . 001 , because they argue that this ensures that the prior contributes little to the posteriors ., However , we propose that this argument is not necessarily always valid ., In particular λi could be expected to be sensitive to priors because it is essentially only fitted to two data points , xi and yi ., Yet , based on approximations of conditional distributions , Tebaldi et al . argued that the gamma distribution with aλ = bλ = 0 . 001 is appropriately vague for the analysis ., For transparency we here follow their approach and investigate the effect of the prior for the simplified model where β = 0 . The mean of the conditional distribution of λi is then approximated by, E ( λi|X0 , X , Y ) ≅aλ+1bλ+12 ( xi−μ˜ ) 2+θ ( yi−ν˜ ) 2 ,, ( 11 ), where μ~ is the conditional mean of the distribution of μ , given by, μ~= ( ∑i=1nλiXi+τ0x0 ) / ( ∑i=1nλi+τ0 ), ( 12 ), and ν~ the corresponding value for v , given by, ν~= ( ∑i=1nλiyi ) / ( ∑i=1nλi ) ., ( 13 ), We stress that Σλi need not sum to one , as might be intuitive when using weights ., As given by Eqs 11 and 12 , the mean of μ and ν only depends on the relative values of λi , but the absolute values influence the width of the distribution , with the variance of the conditional distributions increasing with lower absolute values of λi ( Table 3 ) ., While a low value of aλ certainly ensures little contribution to the numerator in Eq ( 11 ) , it is less evident that a low value for bλ contributes little to the denominator because if xi→μ~ and y→ν~ , the denominator actually approaches bλ ., Hence , to ensure that a low value of bλ can be considered vague such that our posterior is informed primarily by x0 , x and y , we must conclude that |xi-μ~| or |yi-ν~| is clearly separated from zero ., However , if λi≫λj for all i ≠ j and λi≫τ0 , then μ~≈xi and ν~≈yi and nothing in the model prevents this relationship ., In fact , if we consider the gamma prior with shape and scale parameters set to 0 . 001 , the distribution has most of its density near zero , however with a fat tail ( yet exponentially bounded ) that allows for high values ., In the current analysis , this corresponds to the prior belief that the majority of modeling assumptions will have very low precision while a few will have very high ., Under this prior belief , it is expected that for some model i , λi≫λj for all i ≠, j . In the instance where instead τ0≫λi for all i , then μ~≈x0 and the approximation would hold , but we cannot expect that relationship ., As we cannot a priori be sure that the choice of aλ and bλ does not influence our posterior as long as they are arbitrarily small , we performed a prior sensitivity analysis and re-ran the analysis with aλ = bλ = 0 . 01 and aλ = bλ = 0 . 0001 ., We could expect that the sensitivity to priors depends on the difference among modeling assumptions , and we investigate this by analyzing ensembles with little and large discrepancy between assumptions in the ensemble as given by Table 1 . We refer to this as the Non Hierarchical Weighting ( NHW ) method ., If we cannot ensure that the analysis is insensitive to the choice of prior , it implies that our prior beliefs will influence how much different projections contribute to ensemble predictions with the current method ., Using prior beliefs is sometimes desirable , and in section 2 . 6 we consider the situation where we trust some modeling assumptions more than others ., However , it would rarely be the case that we would have substantial expectations that could inform the shape , aλ , and scale , bλ , of the prior for λ ., A potential solution might be to extend the model to include hierarchical priors such that the prior for λi is estimated in the model rather than a priori fixed ., We make a slight change to the parameterization of the prior distribution such that, λi~Gamma ( aλ , aλ/mλ ) ,, ( 14 ), i . e ., specifying the distribution by its mean mλ and shape aλ , which are estimated in the model ., In that way , we move our uncertainty up a level and express our beliefs about the distribution of mλ and aλ , rather than λ ., Using mλ rather than bλ facilitates the specification of a prior for the mean precision parameter that corresponds to the priors previously specified on individual λi ., This parameterization further aids the use of prior beliefs about weights in section 2 . 6 ., While we can never be strictly uninformative in Bayesian analysis , the hierarchical prior can allow for a wide range of plausible mλ and aλ whereas the model presented in section 2 . 4 requires these to be specified explicitly ., This also allows for the concept of “borrowing strength” 54 , such that the distribution of each λi can be indirectly informed by all other precisions via the hierarchical distribution ., This is often beneficial in situations where individual parameters are fitted to a small amount of data 55 , 56 , which clearly is the case for λi here ., To extend Eq ( 10 ) to a hierarchical model , we include hyperpriors such that, aλ~Gamma ( Aaλ , Baλ ), ( 15 ), and, mλ~Gamma ( Amλ , Bmλ ) ., ( 16 ), We performed the corresponding sensitivity analysis for the hierarchical ensemble prediction by applying hyperpriors Aaλ = Baλ = Amλ = Bmλ set to 0 . 01 , 0 . 001 and 0 . 0001 ., We refer to this as hierarchical sensitivity set-up one ., Secondly , we performed a sensitivity analysis , hierarchical sensitivity set-up two , where we fixed the shape parameters Aaλ = Amλ = 0 . 001 and only varied Baλ = Bmλ , again set to either 0 . 01 , 0 . 001 or 0 . 0001 ., We refer to this as model as the Standard Hierarchical Weighting ( SHW ) method ., Using expert opinions may substantially improve predictions 57 , and there are several instances where incorporating prior beliefs that reflect the “trust” in different modeling assumptions could be useful ., For instance , a policymaker might have more trust in one model type over another , and rather than excluding the models that are considered less reliable ( i . e . giving them a priori zero weigh ) , it could be useful to include them , yet with less contribution to the ensemble ., In the case considered here , where modeling assumptions represent most likely , best and worst case in terms of parameterization , we might want to give the “most likely” modeling assumption higher weight ., For the analysis with fixed aλ and bλ , described in section 2 . 4 , we could merely elicit a different scale parameter bλ for each λi , such that modeling assumptions with high trust are given a low value ., However , with the shape parameter aλ set to a low value ( “vague” shape ) , the prior may have little effect on the posterior λi ., Eliciting a high value of aλ would instead result in a posterior that is merely the results of our prior beliefs and we have no foundation for which to elicit some intermediate value ., In order to combine the hierarchical approach with informative priors , we propose a modification of the analysis presented in section 2 . 5 , where the assumption of exchangeability is relaxed in the hierarchical structure with, λi~Gamma ( aλ , aλ/m^λi ) ,, ( 17 ), where m^λi=wimλ and wi indicates the a priori trust in modeling assumption, i . With wi = kwj , the prior distribution of λi will have a mean that is k times that of λj and from Eqs ( 12 ) and ( 13 ) the relationship also implies that before λ is estimated ( i . e . involving the data x0 , x and y ) , the outputs of modeling assumption i will contribute k times as much to μ and v as does assumption, j . To demonstrate the effect that a priori trust in different modeling assumptions can have on the posterior estimates , we consider the case where the best , most likely and worst case scenarios for each of the two varied parameters corresponds to percentile 2 . 5 , 50 and 97 . 5 , respectively , of a normal distribution ., Given that the density at percentiles 2 . 5 and 97 . 5 then is 0 . 15 of that at the mode , we specify wi = 0 . 15 for i = 2 , 3 , 4 and 7 ,, i . e ., for modeling assumptions where one of the varied p

Introduction, Materials and Methods, Results, Discussion

Mathematical models are powerful tools for epidemiology and can be used to compare control actions ., However , different models and model parameterizations may provide different prediction of outcomes ., In other fields of research , ensemble modeling has been used to combine multiple projections ., We explore the possibility of applying such methods to epidemiology by adapting Bayesian techniques developed for climate forecasting ., We exemplify the implementation with single model ensembles based on different parameterizations of the Warwick model run for the 2001 United Kingdom foot and mouth disease outbreak and compare the efficacy of different control actions ., This allows us to investigate the effect that discrepancy among projections based on different modeling assumptions has on the ensemble prediction ., A sensitivity analysis showed that the choice of prior can have a pronounced effect on the posterior estimates of quantities of interest , in particular for ensembles with large discrepancy among projections ., However , by using a hierarchical extension of the method we show that prior sensitivity can be circumvented ., We further extend the method to include a priori beliefs about different modeling assumptions and demonstrate that the effect of this can have different consequences depending on the discrepancy among projections ., We propose that the method is a promising analytical tool for ensemble modeling of disease outbreaks .

Policy decisions in response to emergent disease outbreaks use simulation models to inform the efficiency of different control actions ., However , different projections may be made , depending on the choice of models and parameterizations ., Ensemble modeling offers the ability to combine multiple projections and has been used successfully within other fields of research ., A central issue in ensemble modeling is how to weight the projections when they are combined ., For this purpose , we here adapt and extend a weighting method used in climate forecasting such that it can be used for epidemiological considerations ., We investigate how the method performs by applying it to ensembles of projections for the UK foot and mouth disease outbreak in UK , 2001 ., We conclude that the method is a promising analytical tool for ensemble modeling of disease outbreaks .

journal.pcbi.1006305

A dual regulation mechanism of histidine kinase CheA identified by combining network-dynamics modeling and system-level input-output data

Living systems sense environmental signals and respond by altered behaviors ., Control of behavior is achieved by a myriad of biochemical reactions forming a reaction network , in which some reactions are directly regulated by the sensed signals ., However , exactly which reactions are controlled by a signal and how that control is exerted are incompletely understood ., Experimentally , it is common that only the final outputs can be measured , whereas changes in internal processes remain largely inaccessible ., Given these limitations in our knowledge of the network and in the available data , inferring the underlying regulation mechanisms from systems-level input-output measurements is a major challenge in systems biology , especially for complex reaction networks 1 ., Here , we address this challenge in understanding regulation mechanisms for the specific case of histidine kinase CheA which is a central component in the machinery of bacterial chemotaxis and which is regulated by bacterial chemoreceptors 2–4 ., The first stage of signal transduction in bacterial chemotaxis is from the external stimulus ( a chemical ) to a conformational change in the transmembrane chemoreceptors 5–7 ., The second stage is from the conformation state of the receptors to modulation of phosphorylation of the response regulator CheY 8 , 9 ., The final stage is control by phosphorylated CheY of the bacterial rotary motor and thus the pattern of cell movement 10 ., The second stage involves histidine kinase CheA , which catalyzes phosphoryl transfer between ATP and a histidinyl side chain in its P1 domain 11 , 12 ., That phosphoryl group is then transferred to an aspartyl residue on CheY ., Phosphoryl transfer from ATP to His to Asp is the hallmark of two-component regulatory systems , which are prevalent in microorganisms 13 ., Phosphoryl transfer from ATP to CheY involves steps internal to CheA 14 ., These are chemical reactions involving ATP , CheY , and three domains of CheA: P1 ( phosphoryl-accepting ) , P2 ( CheY-binding ) , and P3P4P5 ( dimerization , catalytic and receptor-coupling , respectively ) 15 , 16 ., These components define chemical states connected by bi-directional ( forward and backward ) reactions , forming a reaction network ., The P3P4P5 unit plays a central role in the network , as the enzyme that catalyzes phosphoryl transfer between ATP and P1 ., For enzymatic reactions with one enzyme Etot and one varied substrate S , experimentally measured reaction rates are typically fit to the Michaelis-Menten equation ,, v = k cat S E tot S K m S + S ., ( 1 ), For chemotaxis phosphoryl transfer , this straightforward analysis has been used to determine how different receptor states regulate the Michaelis-Menten constant K m S and the catalytic rate constant k cat S 17 ., However , such analysis cannot reveal underlying regulatory mechanisms 18 ., To determine which steps of the reaction network are regulated and how they are regulated is essential to understand the signaling mechanism ., However , it is also an extremely challenging task because it is difficult , if not impossible , to probe experimentally each individual reaction in the network separately ., In this paper , we investigate the regulatory mechanism of kinase CheA by modeling the kinetics of the entire enzymatic reaction network using different hypotheses of regulation ., Models were fit to an entire set of experimental data with the aim of determining which hypotheses were consistent with the data ., The best-fitting models suggested experiments to distinguish among them ., This systematic approach , demonstrated here for CheA regulation , should be applicable to investigations of other complex biochemical networks ., To explain this full set of data within a coherent framework , we developed a simple enzymatic network model to describe dynamics of the enzyme in all possible states in combination with its two substrates/products ( ATP/ADP , P1/P1P ) ., The enzyme has two binding sites , one for ATP or ADP and other for P1 or P1P ., Each binding site can be in three states: empty , occupied by substrate or by product ., The combination of these occupancy states results in 3 × 3 = 9 enzyme configurations shown in Fig 1 . The transitions from one state to another form the enzymatic reaction network ., The association/dissociation dynamics of substrate S ( = ATP , P1 ) to the enzyme are described by the dissociation rate constant , k off S , and by the equilibrium dissociation constant , K d S , which identify the equilibrium properties of the binding process ., For convenience , we defined the on rate as, ω S = k off S S K d S ., ( 3 ), The experimental design in which the reactions proceeded for only a brief time before sampling allowed us to assume that ATP and P1 were constant and that ADP ≈ P1P ≈ 0 . This latter assumption leads to ωADP ≈ ωP1P ≈ 0 , represented by gray colored arrows in Fig 1 . The phosphoryl group transfer rate constant for transfer from ATP to P1 is k f P . The reverse rate constant k r P describes the opposite transition ., The ratio between these two rate constants, GP=krPkfP=ATP·E·P1ADP·E·P1P|Isol . eq ., ( 4 ), defines the isolated equilibrium between the two states ( ATP · E · P1 and ADP · E · P1P ) ., The equilibrium properties depend only on the difference of free energy between the two enzyme configurations , which is given by kBT ln GP with kBT the thermal energy unit ., Details of the mathematical formulation of the enzymatic dynamics illustrated in Fig 1 are given in the S1 Appendix ., Besides defining the chemical reaction network that connects different states of the enzyme , another important ingredient of the model was to specify which reactions ( links in the network ) are regulated by the receptor and how they are regulated ., Let us characterize the strength of the receptor’s regulatory activity by 0 ≤ σ ≤ 1 , σ = 0 , 1 correspond to the minimum and the maximum activities respectively ., In our network model of the enzyme kinetics , there are three possible steps that the receptor activity ( σ ) could affect: association/dissociation of ATP/ADP , association/dissociation of P1/P1P , and the phosphoryl transfer between ATP and P1 ., Regulations of different reactions are labeled by different colors of the reaction arrows in Fig 1 . In this paper , we consider all three possibilities and their combinations to identify what are the minimal sets of regulations needed to explain all the experimental data 17 ., The exact nature of the regulation determines how σ affects the reaction rate kn for the nth reaction ., Here , we consider the simple case in which the enzyme has two conformations ( active and inactive ) and the receptor controls the fraction of time the enzyme spends in each conformation ., We further assume switching between the two enzyme conformations happens at a timescale much faster than other chemical reactions ., As a result , the total reaction rates are weighted averages of the “bare” rates in each enzyme state and can be expressed as simple functions of σ depending on the nature of the regulation ., Specifically , σ dependence is linear if the receptor only affects the kinetic rate constants without changing equilibrium properties of the enzyme , but it is a linear rational function if the receptor also changes the equilibrium properties of the enzyme such as the binding energies ( see S2 Appendix ) ., We used our model together with the input-output data to determine the nature of the regulation ., For each specific model using a specific hypothesis Hi , we find the values of the biochemical parameters p → and the receptor activities σ → that minimize the mean error function χ2 over all experiments defined as:, χ 2 ( H i ) = min p → , σ → 1 N ∑ j = 1 10 N j χ j 2 ( p → , σ j | H i ) , ( 5 ), where j ∈ 1 , 10 labeled all the 10 individual receptor states characterized by the membrane preparation ( “v” for vesicle , “d” for nanodisc ) , receptor methylation level ( EEEE , QEQE , QQQQ ) , and the ligand ( aspartate ) concentration ( in μM ) ., The total number of data points was N = ∑ j = 1 10 N j , where Nj was the number of experimental points in each state ( between 12–14 data points ) ., The error ( loss ) function χ j 2 for experiment j was defined as, χ j 2 ≡ 1 max j ( y ) ∑ iin statej ( y i −f ( x →i ) ) 2 y i , ( 6 ), where the difference ( residue ) between the data value and the model fit , ( y i - f ( x → i ) ) , was scaled by the geometric mean of the experimental value yi and the maximum value maxj ( y ) of the entire experiment j ., We used this definition of error function to avoid giving too much weights to data points with large measured values ., All model parameters fell into two categories ., The first category of parameters p → included three kinetic rate constants ( k off ATP , k off P1 , k f P ) and three equilibrium constants ( K d ATP , K d P1 , GP ) , which described the basic chemical reactions: association/dissociation of P1 and ATP to the enzyme and phosphoryl transfer between bound P1 and ATP ., These “bare” biochemical parameters are the same in all experiments ., The second category of parameters was the experiment-specific receptor activities σj , which could modulate the biochemical rates for each different receptor state j ( ∈1 , 10 ) ., We set the activity of the most active receptor to be unity ( σ10 = 1 ) to set the scale of the receptor activities , and the number of activity parameters was reduced to 9 ., Thus there are 15 parameters in our model , which is used to fit all the 20 experimental response curves simultaneously ., We used a non-linear solver ( Levenberg-Marquardt algorithm of the function curve_fit from the package SciPy 20 ) to find the set of parameters p → * and σ → * that minimizes the mean error χ2 defined by Eq 5 ., For statistics based data fitting where the underlying mechanism is unknown , the balance between improvement in the fitting and the number of parameters may be assessed by information criteria such as the Bayesian 21 or Akaike 22 methods ., Here , the structure of our model and the model parameters were all based on known underlying biochemical processes ., Therefore , the number of parameters Ntotal in our models does not vary widely as shown in Table 1 and Ntotal can not be used effectively to differentiate different models/hypotheses ., Instead , by following the approach used in the Global Kinetic Explorer 23 , we studied the sensitivity of the error function χ2 with respect to changes in each model parameter and the correlations between different parameters ., The error function landscape in the parameter space was investigated by fixing a particular parameter pi to a value around its optimal value p i = x i p i * and minimizing χ2 by varying all other parameters pk , k ≠ i and all activities σ →:, χ ˜ 2 ( x i ) = min p k , k ≠ i , σ → 1 N ∑ j = 1 10 N j χ j 2 ( p i ) ., ( 7 ), The dependence of χ ˜ 2 ( x i ) on all individual parameters xi in model 6 were shown in Fig 2, ( a ) , where all the curves have the same minimum at xi = 1 . The larger the curvature at the minimum , the more sensitive is the fitting to the change of that parameter ., From the second order derivative at the minimum , we can determine a range for the fitted parameter as given in Tables 2 and 3 ., The sensitivity analysis was essential in identifying underconstrained parameters ., In particular , when there is a fast reaction with a timescale that is much shorter than the experiment time and the other relevant timescales in the system , the exact value of the rate constant for this ultra fast reaction can not be determined uniquely , only a lower bound can be established ., For example , in our model H6 ( see Table 1 ) , if k off P1 were ( unknowingly ) treated as a fitting parameter , the sensitivity analysis showed only a weak dependence on k off P1 as long as it is bigger than a certain value as shown in Fig 2, ( b ) ., This excessive degree of freedom also impairs the fitting algorithm , which generates the irregularities seen in Fig 2, ( b ) ., To avoid fitting an underconstrained parameter , we fixed k off P1 to be a large rate constant k off P1 = 100 s - 1 in model H6 ., The parameters used in our models came from the underlying biochemical reactions and they were not orthogonal in the error function landscape ., As shown in Fig 2, ( c ) , when we changed k f P from its optimal value and allowed the other parameters to vary to minimize χ2 , the optimal values of k off ATP and k off P1 also changed roughly proportionally to k f P while other parameters only showed weak dependence on k f P . The correlation between two parameters can be characterized by the linear proportionality constant between the two parameters ., In Fig 2, ( d ) , we show all the pair-wise linear correlation constant in a matrix ., These correlations can be understood intuitively ., Since k f P and k off ATP are slow reactions in model H6 ( see Table A in S4 Appendix ) , they must change together to preserve the qualitative behavior ., An increase in the forward phosphoryl transfer reaction rate constant k f P can be partially compensated by an increase in GP , which enhances the reverse phosphorus transfer reaction ., These partial compensation mechanisms also explained the lower precision in these parameters as seen in Fig 2, ( a ) as compared with the parameters K d ATP , K d P1 , an K d / D P1 that have exclusive control of certain parts of the network ., Additional parameters were added to the model based on realistic biological considerations , e . g . , the values of Kd could be different for nanodiscs and vesicles; or possible regulation hypothesis , e . g . , there could be a residual kinase activity in the inactive state ., When additional parameters allowed a better fitting to the data , a well defined minimum of χ2 emerged in the error function analysis ., This was the case when we considered different values of Kd for nanodiscs and vesicles ( model H8 in Table 1 ) as shown in Fig . B in S6 Appendix ., The number of parameters and the value of χ2 obtained from a representative subset of models we studied are shown at Table 1 . Among the multitude of models we tried , these are the ones with roughly the lowest χ2 for each number of parameters , with the exception of H1 ., It clearly demonstrates that a model does not necessarily fit the data better just because it has more parameters ., As shown in Fig 3, ( a ) , none of the three single regulation hypotheses ( H1 , H2 , and H3 ) , i . e . , regulating the ATP or P1 binding or the phosphoryl transfer rate constants fit all the experimental data ., To our surprise , the single regulation of k off ATP ( H3 ) was much better than the other two single regulation mechanisms ., The worst performing single regulation hypothesis was regulating P1 dissociation k off P1 ( model H1 ) , where the errors from several experiments such as VEEEE10 , vQEQE5 , vQEQE100 , and vQQQQ0 were large ., For the model with single regulation of the phosphotransfer rate k f P ( model H2 ) , the fitting of vQQQQ0 improved ., However , the error for the vEEEE10 experiment was still large ., The reason for the large fitting errors for receptor states like vEEEE10 is due to their lower activities than others ., It is the extreme low activity receptor state that shows the largest difference between models with and without certain regulation by receptor activity ., The detailed reason for this misfit can be understood by comparing the model results with experimental data directly ., As shown in Fig 3, ( b ) , the maximum kinase rate and the half-maximum ATP concentration for vEEEE10 are both higher in the experiment than in the model ., Thus an improvement in fitting this curve seems to suggest an additional regulation of k off ATP ., Based on results from all single regulation models , we next tried to combine the different regulations ., We found that there was a general reduction of errors across most experiments by having k off ATP regulation combined with a regulation of either k f P ( from H2 to H4 ) or k off P1 ( from H1 to H5 ) without introducing any additional parameters in our model ., The decomposed fitting errors for these two successful dual regulation mechanisms are shown in Fig 3, ( a ) ( models H4 and H5 in the second row of the upper legend ) ., However , the dual regulation of k f P and k off P1 did not improve the fitting and resulted in a larger error χ2 = 0 . 132 ., For a given reversible chemical reaction between two states , the receptor activity ( σ ) can change the energy barrier between the two states and thus change the kinetic rate constants by the same factor ( linearly proportional to σ ) without changing their ratio , i . e . , the equilibrium constants K d ATP , K d P1 , and GP ., This was the situation we considered in most of our study ., However , we also considered the more general cases in which the receptor activity changed the free energy difference between the two states leading to different dissociation constants for the active ( σ = 1 ) and inactive ( σ = 0 ) receptors and a more complicated ( linear rational function ) dependence of the forward and backward rate constants on σ ( see S2 Appendix ) ., With this new degree of freedom , only slightly improved fittings were achieved as shown in Fig 3, ( a ) ( the fourth row in the legend ) for a model H8 with residual activity in P1 binding ., We also allowed this new degree of freedom in the single regulation cases , but the fittings did not seem to improve much if at all ( see Fig . A in S2 Appendix ) ., Our results suggest that receptors mainly regulate the kinetic rates by controlling the energy barrier between two states without changing their free energy difference ., In both dual-regulation mechanisms , our model indicated that for cases with the same receptor modification state ( QEQE ) and the same aspartate concentrations ( 0 , 5μM ) , the receptor activities were larger in nanodiscs than in vesicles by ∼60–90% ( see Table 3 and Table A in S4 Appendix ) ., The differences could reflect higher activity of receptors removed from the heterogeneous native membrane or a difference between activity of the small clusters of signaling complexes that form when participating receptors are native membrane vesicles 24 and individual core complexes are constructed using receptors inserted in nanodiscs 25 ., We also considered the possibilities that there were different values of k off P1 , K d P1 , K d ATP , or k off ATP for vesicles and nanodiscs ., We found that a modest improvement in fitting could be achieved by having different values of K d P1 for nanodisc and vesicles ., These hypotheses ( H6 and H7 ) are shown in the third row in the legend of Fig 3, ( a ) , details of these models can be found in Table 1 ., Table 2 and Table A in S4 Appendix show the parameters of the two dual-regulation models ., In both models ( H6 and H7 ) , our study suggested a higher value of K d P1 for nanodiscs than that for vesicles by about 50% ., The actual fitting of this model ( H7 ) to the experimental data is given in Fig 4 ., We also explored the possibility of the binding of one substrate depending on the presence of the other substrate as investigated in 26 , 27 and the possibility of K d S and k off S being different for for P1 and P1P or for ATP and ADP as proposed in 28 ., However , including these possibilities did not improve the fitting of the available data ., Further experiments are needed to explore these more detailed hypotheses ., Dynamics of the enzymatic network are determined by the transition rate constants between different states in the network ., These different rate constants give rise to different time scales , which can be regulated by receptor activity ., To demonstrate the importance of time scales , in Fig 6 we plot the apparent phosphoryl transfer rate constant as a function of time for the least active receptor EEEE ., We followed the same procedure as in the experiments that generated the data we have analyzed , i . e . pre-mixing P1 with the enzyme and then adding ATP at time t = 0 ., Depending on details of regulation by receptors , the apparent phosphorylation rate constant could either decrease after an initial fast surge ( the blue lines ) or increase from zero gradually before converging to its steady state value ( the orange lines ) ., The convergence to the steady state is characterized by the relaxation time , τ ., Precise calculation of the relaxation time is presented in the SM , but it can be estimated as the inverse of the lowest reaction rate constant ., From the inverse of k off ATP from Tables 2 and 3 , the longest relaxation times were observed for kinase control by EEEE receptors , estimated as 54 s for Asp = 0 and 362 s for Asp = 10 μM ., This means that the average phosphorylation rates determined in the experiments with Δt = 15 s , 60 s are not steady state rates ., We have taken this time-dependent effect into account explicitly in all our model fittings ., Since Michaelis-Menten ( MM ) analysis of enzyme kinetic data assumes a steady state in which the concentration of enzyme-substrate complex is constant and thus substrate binding to the enzyme is equilibrated , if this condition is not met the MM parameters obtained from the analysis will be incorrect ., This can be a problem for analysis of very low activity receptors which generate very low kinase catalytic rate constants ( kcat ) ., However , the effective ( phenomenological ) MM fitting parameters provide useful information by identifying maximum reaction rates and thus rate constants , as well as the substrate concentration , Km , at which the reaction rate is half maximum ., We determined the values of these effective MM parameters for curves obtained from our model and compared them with those obtained from direct fitting of the experimental data with the MM equation ., There was good agreement between MM parameters obtained in the two ways ., However , our analysis also illustrated the potential errors in determining in MM parameters in conditions in which reaction rate constants are very low and thus care must be taken to insure that experimental samples are taken after sufficient time for the MM steady-state assumption to be valid ( Fig 7 ) ., Specifically , long relaxation times for low values of k c a t S lead to effective Km values that depend on the measurement time Δt ( see Fig 7 ) , because the MM requirement for a steady state is not met ., The rich dynamics of the enzymatic reaction network before it reaches its steady state suggest future experiments that can be used to distinguish the remaining hypotheses ., By using our model , we can determine the experimental conditions in which the different hypotheses lead to different dynamic behaviors ., The difference in dynamic behaviors is most prominent for the less active receptors ( EEEE ) where the relaxation times are long ., Fig 6 illustrates how two of the best fitting models ( H6 and H7 ) , both of which explain the existing experimental data at a particular time Δt = 60s , lead to distinctive phosphorylation time courses due to their different rate limiting steps ., Model H8 is slightly better than H7 , but has an extra parameter that does not significantly alter the dynamics ., For the sake of simplicity we will use H7 ., In the dual regulation model ( H7 ) of ATP and P1 binding , both P1 and ATP bindings are the rate limiting steps for low receptor activity states such as the EEEE receptor with a aspartate concentration Asp = 10 μM ., In the experiments that generated the data we analyzed , P1 was mixed with the enzyme prior to initiation of the reaction by addition of ATP thus bypassing this limiting step , the other limiting step is lifted by a large ATP concentration ATP = 10 mM ., As a result , the phosphorylation rate rises quickly to its maximum before decreasing to its steady state value as shown in Fig 6 ( blue lines ) ., In the other dual regulation model ( H6 ) of ATP binding and phosphoryl transfer , a much slower initial increase in the phosphorylation rate is predicted ( orange lines in Fig 6 ) because the rate limiting phosphoryl transfer reaction can not be bypassed ., The full time-dependent phosphorylation rates shown in Fig 6 represent quantitative predictions that can be tested in future experiments to verify our model and to distinguish the different dual regulation mechanisms ( H6 versus H7 ) ., Our model also shows that the transient phosphorylation kinetics depend on the initial incubation process ( premixing P1 or ATP with the enzyme ) , which can also be tested by future experiments ., In particular , the phosphoryl transfer rates can be measured continuously ( at least at multiple times ) in the time window of 0–100 s ., If the time dependence follows that of the blue ( red ) line shown in Fig 6 , it would indicate that the underlying mechanism is H7 ( H6 ) ., See S6 Appendix for details ., In general , understanding microscopic mechanisms in biological systems is challenging given the complexity of the underlying processes and the difficulty in measuring individual reactions ., Here , we show that combining modeling of the dynamics of the whole reaction network with quantitative system level “input-output” measurements provides a powerful tool to address this challenge , as demonstrated here in the case of kinase CheA regulation ., This systems-biology approach , which includes the development of a mechanistic network model based on key underlying biochemical reactions and searching the hypothesis space by fitting a large body of input-output data to the model , should be generally applicable to the study of other biological regulatory systems .

Introduction, Models and methods, Results, Discussion

It is challenging to decipher molecular mechanisms in biological systems from system-level input-output data , especially for complex processes that involve interactions among multiple components ., We addressed this general problem for the bacterial histidine kinase CheA , the activity of which is regulated in chemotaxis signaling complexes by bacterial chemoreceptors ., We developed a general network model to describe the dynamics of the system , treating the receptor complex with coupling protein CheW and the P3P4P5 domains of kinase CheA as a regulated enzyme with two substrates , ATP and P1 , the phosphoryl-accepting domain of CheA ., Our simple network model allowed us to search hypothesis space systematically ., For different and progressively more complex regulation schemes , we fit our models to a large set of input-output data with the aim of identifying the simplest possible regulation mechanisms consistent with the data ., Our modeling and analysis revealed novel dual regulation mechanisms in which receptor activity regulated ATP binding plus one other process , either P1 binding or phosphoryl transfer between P1 and ATP ., Strikingly , in our models receptor control affected the kinetic rate constants of substrate association and dissociation equally and thus did not alter the respective equilibrium constants ., We suggest experiments that could distinguish between the two dual-regulation mechanisms ., This systems-biology approach of combining modeling and a large input-output dataset should be applicable for studying other complex biological processes .

In complex biological systems , it is often difficult to determine which steps in the underlying biochemical network are regulated by the signal by using direct experimental measurements alone ., In this paper , we tackled this general problem in the case of the kinase activity of the multi-domain histidine kinase CheA ., We developed a quantitative reaction network model to describe the CheA enzyme kinetics by considering all the key reaction steps explicitly ., We used this general model with different regulation schemes of progressively increasing complexities to fit a large input-output dataset ., Our modeling revealed novel dual regulation mechanisms in which receptor activity regulated two independent reactions in the network including the ATP binding reaction that was previously unsuspected ., Through our quantitative analysis , we found that receptors affected the kinetic rate constants of substrate association and dissociation equally and thus did not alter the respective equilibrium constants ., Testable predictions of the kinase activity dynamics are made from our models to further distinguish the different dual regulation mechanisms ., Our study shows that combining modeling kinetics of the reaction network and input-output data can help reveal the underlying regulation mechanism in complex networks where probing individual reaction is impossible .

phosphorylation, cell motility, vesicles, enzymes, enzymology, chemical equilibrium, membrane receptor signaling, network analysis, cellular structures and organelles, enzyme chemistry, physical chemistry, computer and information sciences, proteins, enzyme regulation, chemistry, chemotaxis, biochemistry, signal transduction, cell biology, post-translational modification, biology and life sciences, physical sciences, cell signaling

journal.ppat.1006764

Exosomes serve as novel modes of tick-borne flavivirus transmission from arthropod to human cells and facilitates dissemination of viral RNA and proteins to the vertebrate neuronal cells

Exosomes are small membranous extracellular microvesicles ( 30 to 250 nm in diameter ) of endocytic origin formed in late endosomal compartments ( as multivesicular bodies; MVBs ) of several different cell types 1–5 ., Initially , exosomes were considered as garbage bins to discard the unwanted cellular or molecular components or membranous proteins from reticulocytes 6–9 ., Other studies have suggested that exosomes are mere cell debris or apoptotic blebs and signs of cell death 10–12 ., Recently , the role of exosomes has been highlighted in important medical research on cancer and autoimmune diseases and they are now recognized as novel therapeutic targets for neurological disorders such as Parkinson’s disease 11 , 13–16 ., Over the past 10 years , exosomes have been given potential biological significance by identifying a variety of their specific roles 3 , 5 , 11 , 17–20 ., Exosomes derived from several different cells have been shown to function as signaling related vesicles , transporting cell-specific collections of several proteins , lipids and nucleic acids such as DNA , RNA and microRNA 12 , 20–28 ., Exosomes are released into circulation after the fusion with the plasma membrane and these vesicles serve as mediators of molecular transmission 3 , 10 , 18 , 29 ., Cell-derived exosomes have been shown to be important modes of intercellular communication and as transmitters of information over longer distances for e . g . , between different tissues or multiple organs 2 , 15 , 27 , 30 , 31 ., Studies have also shown that exosomes are vehicles of transmission for a variety of microorganisms and that some pathogens uses exosomes to manipulate their environments 10 , 15 , 32–34 ., As an example , malaria parasites , Plasmodium falciparum , uses exosomes for communication between infected red blood cells 35 ., Hepatitis C virus ( HCV ) , an enveloped RNA virus , associates with exosomes isolated from cell culture supernatants and from infected patients 36 , 37 ., Recent findings of HCV transmission through hepatic exosomes establish infection provides new insight into hepatitis drug discovery 38 , 39 ., Exosomes also function in the transfer of immuno-stimulatory viral RNA from HCV-infected cells to co-cultured plasmacytoid dendritic cells 32 ., In addition , exosomes facilitate receptor-independent transmission of replication-competent HCV viral RNA that was found to be in complex with Ago2-miR122-HSP90 in HCV-infected individuals or infected hepatocytes 36 ., Interestingly , exosomes have been shown to play dual roles in transmitting Hepatitis A virus ( HAV ) and HCV , thereby evading antibody-mediated immune responses 40 ., It has been demonstrated that Toll-like receptor 3 ( TLR-3 ) activated macrophages release exosomes containing anti-HCV micro ( miRNA ) -29 family members that suggest a novel antiviral mechanism against HCV infections 41 ., Herpes Simplex-1 virus and Epstein-Barr virus also use exosomes for transmission 42 , 43 ., Several studies have suggested exosomes as important players in HIV-1 pathogenesis 33 , 34 , 44 ., HIV Nef protein secreted in exosomes has been shown to trigger apoptosis in CD4+ T cells and the Gag p17 coding RNA is also targeted to the exosomes 45 , 46 ., HIV-infected cell-derived exosomes have been shown to contain the TAR ( Trans-Activation Response Element ) miRNA that facilitates production of pro-inflammatory cytokines 47 , 48 ., Moreover , a recent but very highlighting study showed that exosomes from uninfected cells activates the transcription of latent HIV-1 49 ., Ixodes ticks transmit several viruses belonging to the family Flaviviridae such as tick-borne encephalitis virus ( TBEV ) , Powassan virus ( POWV ) and Langat virus ( LGTV ) 50–53 ., LGTV is considered as a model biosafety level 2 ( BSL2 ) pathogen to study pathogenesis of TBEV , due to its significant genome homology with the later ., Transmission modes of these arthropod-borne flaviviruses ( with positive sense single-stranded RNA ) are poorly understood 37 , 54 ., Our study shows for the first time that exosomes facilitate transmission of flavivirus RNA and proteins from arthropod to human cells ., We have demonstrated that cells from the medically important vector tick , Ixodes scapularis , secretes exosomes that mediate transmission of tick-borne LGTV RNA and proteins from arthropod to human ., Our study shows the presence of abundant amounts of LGTV RNA and proteins in exosomes isolated from arthropod and neuronal cells ., We also found that LGTV-infected tick cell-derived exosomes were capable of transmigrating and infecting naïve human skin keratinocytes ( the initial barrier lining the human cells that comes in contact during bites from infected ticks ) and human vascular endothelial cells ( that comes in contact during arthropod blood feeding ) ., Our data show that vertebrate exosomes mediate transmission of tick-borne LGTV RNA and proteins from infected-brain microvascular endothelial cells ( a component of the blood-brain barrier; BBB ) to neuronal cells ., In addition , we have demonstrated that exosomes containing tick-borne LGTV and mosquito-borne West Nile virus ( WNV ) facilitate transmission of viral RNA and proteins from one neuronal cell to others suggesting their novel role in neuropathogenesis ., Dihydrochloride hydrate , GW4869 ( a selective inhibitor for neutral sphingomyelinase; N-SMase , an enzyme that regulates production and release of exosomes ) , reduced LGTV loads in exosomes and inhibited the transmission of LGTV RNA and proteins in both arthropod and vertebrate host cells ., Overall , our study suggests that exosomes are not only the mediators for transmission of arthropod-borne flavivirus RNA and proteins from arthropod to the vertebrate host , but also facilitate dissemination of these infectious RNA and proteins within the vertebrate host , including crossing of BBB cells and allowing neuroinvasion and neuropathogenesis in the Central Nervous System ( CNS ) ., Despite the significance of ticks as important medical vectors , we know little about the transmission modes of tick-borne viruses and other tick-borne pathogens to the vertebrate host ., We first analyzed whether tick cells secrete extracellular vesicles ( EVs ) and exosomes and if tick-borne flaviviruses use those exosomes as modes of pathogen transmission ., LGTV , a flavivirus closely related to TBEV , readily infected Ixodes scapularis ISE6 tick cells , with increased viremia at 72 h post-infection ( p . i . ) ( S1A Fig ) , similar to the viral infection kinetics observed in Vero cells ( S1B Fig ) ., We selected 72 h p . i . as the time point for the isolation of exosomes from tick cells due to the higher viral loads ., First , we isolated exosomes by density gradient centrifugation technique using OptiPrep ( DG-Exo-iso ) as described in 55 ., This isolation method used in our settings with a floor ultracentrifuge unit is shown as a schematic representation in ( S1C Fig ) ., Exosomes were also independently isolated by differential ultracentrifugation with slight modifications and longer spin times for 155 minutes ( S2 Fig ) 22 , 29 , 32 , 56 , 57 ., We also isolated arthropod-derived exosomes using commercially available exosome isolation reagent following manufacturers instructions ( Invitrogen/ThermoScientific ) ., Notably , all preparations contained 30 to 200 nm vesicles and these techniques have been used extensively in several studies ., Cryo- Electron Microscopy ( cryo-EM ) performed on tick cell-derived exosomal fractions showed the presence of purified arthropod exosomes with the size range of 30 to 200 nm in diameter ( Fig 1A ) , similar to exosomes isolated from mammalian cells 1–3 ., Exosomes isolated from arthropod cells showed a heterogenous population of vesicles in the cryo-EM analysis ., In order to understand such heterogeneity in exosome populations , we did quantitative analysis using images collected from both uninfected and LGTV-infected tick cell-derived exosomes ., We noted that majority of the exosomes were of sizes between 50–100 nm in both uninfected and infected groups ( Fig 1B and 1C ) ., However , exosomes of other sizes 100–150 and 150–200 nm were evenly distributed in infected group when compared to the uninfected group ( Fig 1B and 1C ) ., Fewer vesicles from sizes of 200–250 nm were slightly more in uninfected ( 10 . 1% ) in comparison to the infected group ( 6 . 5% ) ., The large exosomes were very few and were from 0–1 . 5% in both the uninfected and infected groups ( Fig 1B and 1C ) ., Counting of exosomes per image showed higher number of exosomes in LGTV-infected ( n = 14 ) in comparison to the uninfected ( n = 27 ) group ( Fig 1D ) ., This data suggested that LGTV-infection ( 72 h p . i . ) might enhance the production and/or release of exosomes ., The OptiPrep ( DG-Exo-iso ) method yielded purified exosomes in six different fractions ., Immunoblotting analysis ( with highly cross-reactive 4G2 monoclonal antibody that recognizes the viral Envelope ( E ) - protein ) of these fractions ( 20 μl ) showed presence of LGTV E-protein in all six fractions but enriched amounts of E-protein were present in fractions four and five in comparison to the other fractions ( Fig 1E ) ., These results correlated with the size analysis data ( Fig 1B , 1C and 1D ) ., Enhanced detection of LGTV-E protein in fractions four may correspond to the 50–100 nm ( fraction, 4 ) size exosomes that are highly populated ( Fig 1E ) ., As expected , we did not detected E-protein in the fractions from uninfected control ., Total cell lysates ( 20 μg ) from uninfected and LGTV-infected groups were used as internal controls to compare the amounts of E-protein detected in the 20 μl of different fractions used ( Fig 1E ) ., The PonceauS images showing the protein profile serve as control ( Fig 1E ) ., Quantitative Real-Time PCR ( QRT-PCR ) analysis revealed presence of LGTV total mRNA in exosomes isolated from infected tick cells ( Fig 1F ) ., The copy numbers of viral RNA in exosomes derived from LGTV-infected ( 72 h p . i . ) tick cells is shown in ( S3A Fig ) ., In addition , we also determined the presence of both positive and negative sense LGTV RNA strands in tick cell-derived exosomes ( Fig 1G ) ., LGTV mRNA was also evident in exosomes from tick cells cultured and infected in exosome-free FBS medium ( with no cross-contaminating bovine exosomes present in regular commercial FBS ) , further suggesting the presence of viral RNA in tick cell-derived exosomes ( S3B Fig ) ., Presence of LGTV E-protein in tick cell-derived exosomes was further recognized by SDS-PAGE followed by immunoblotting with 4G2 antibody ( Fig 1H ) ., Higher E-protein loads were detected ( at ~50kDa ) in total cell lysates in comparison to exosomal preparations ( Fig 1H ) ., Immunoblotting with monoclonal anti-Langat virus NS1 ( Clone 6E11 ) antibody ( obtained from BEI resources ) also showed the presence of NS1 in both tick cell-derived exosomes and total cell lysates ( Fig 1H ) ., Although , higher NS1 protein loads were evident in total lysates , but the presence of NS1 in tick-cell derived exosomes ( Fig 1H ) further confirmed that these arthropod exosomes contain LGTV proteins ., Remarkably , we also detected the presence of tick HSP70 ( heat-shock cognate protein 70 , a specific exosomal marker in mammalian cells ) in exosomal lysates ( Fig 1H ) ., No differences were noted in HSP70 loads between uninfected and infected exosomal lysates ( Fig 1H ) ., Presumably due to low amount in cell lysates , no HSP70 was detected in the tested condition ( Fig 1H ) ., Total protein lysates prepared from same batch of uninfected or LGTV-infected tick cell-derived exosomes or from whole tick cells served as loading control for all immunoblots ( Fig 1H ) ., It was noted that some of the bands in the total protein profile gel were enhanced in LGTV-infected tick exosome lysates in comparison to the uninfected controls ( Fig 1H ) ., Furthermore , native-PAGE followed by immunoblotting with 4G2 antibody , showed enhanced levels of LGTV E-protein ( at <250kDa; in native state ) in exosomes treated ( 30 min , RT ) with Triton-X-100 ( a detergent that lyses the exosomal lipid bilayer membranes ) in comparison to the exosomes treated for three rounds of freeze-thaw cycle ( 1 h , at -80°C ) or the untreated exosomes held at 4°C ( Fig 1I ) ., Total protein lysates prepared from uninfected or LGTV-infected tick cell-derived exosomes with similar treatments served as controls in this immunoblotting analysis ( Fig 1I ) ., Detection of LGTV E-protein inside exosomes ( but not outside in the PBS suspensions ) was further analyzed by E-protein-4G2-antibody-beads binding assay as described in methods ., No significant ( P>0 . 05 ) differences in viral loads were observed in LGTV-infected ( 72 h p . i . ) exosome samples that were either untreated or treated with 4G2 antibody ( that binds to LGTV E-protein ) or relevant isotype control antibody ( Fig 1J ) ., Similar results were obtained with LGTV-infected exosomal preparations derived from GW4869 inhibitor treated tick cells collected at 72 h p . i . , ( Fig 1J ) ., Native-PAGE and the beads assay clearly suggest that exosomes contain viral RNA and proteins inside exosomes ., To further evaluate if viral E-protein is indeed ( totally ) inside the exosomes , we performed the protease-resistance assay with Proteinase K that generally digest proteins in all biological samples ., We found that treatment with Proteinase K ( 0 . 5 μg/μl or 50 μg/ml , 15 min at 37 ºC ) at typical and suggested working concentrations ( 50–100 μg/ml ) digested all proteins ( S3C Fig ) ., We detected E-protein in untreated infected samples but not in treated infected samples ., Uninfected samples either treated or untreated served as internal controls ( S3C Fig ) ., The Ponceau S stained blot showed no proteins in infected or uninfected proteinase K-treated samples ( S3C Fig ) ., During isolation of tick exosomes , pellet fraction ( containing exosomes ) and supernatant fraction ( generated after pelleting exosomes and before PBS wash; See S2 Fig ) was tested in plaque assays to determine infectivity and replication of viral RNA and titers as described in methods ., Plaque assays performed with the tick cell-derived exosome pellet fractions yielded plaques at dilutions of 1:10 and 1:100 that were too numerous to count , and around 20–22 plaques at a dilution of 1:1000 ( Fig 1K and S3D and S3E Fig ) ., No plaques were detected in plates where Vero cells were treated with the supernatant fractions at any tested dilution ( Fig 1K and S3D and S3E Fig ) ., Plaque assays indicated the presence of infectious viral RNA or proteins in LGTV-infected exosomes that resulted in high loads of LGTV in Vero cells ( 6 . 6 x 104 pfu/ml ) and increased formation of viral plaques ., Plaque assays further confirmed that tick cell-derived exosomes contain LGTV RNA and proteins capable of replication and forming viable plaques that are highly infectious to mammalian cells ( Fig 1K and S3D and S3E Fig ) ., No detection of viral plaques in the supernatant fractions suggests presence of abundant amounts of LGTV RNA and proteins in exosomes ( Fig 1K and S3D and S3E Fig ) ., Overall , these results suggest that majority of the LGTV RNA and proteins exit tick cells via exosomes and that exosomes could mediate transmission of these and possibly the other closely related viruses such as TBEV and POWV ., As tick-borne viruses ( including TBEV , LGTV and Powassan virus ) are transmitted by an infected tick bite to the vertebrate hosts , we tested whether exosomes isolated from LGTV-infected tick cells are infectious to human cells ., In an infection kinetics assay , LGTV readily infected human keratinocytes ( HaCaT cells ) at all tested time points ( 24 , 48 and 72 h p . i . ) and there were no changes in viral loads at different times p . i . ( S3F Fig ) ., Infection of HaCaT cells with exosome fraction prepared from LGTV-infected tick cells ( 72 h p . i . ) showed significantly ( P<0 . 05 ) increased levels of viral loads at 72 h p . i . in comparison to HaCaT cells treated with supernatant fractions prepared from 72 h post-infected-tick cells ( Fig 1L ) ., Tick cell-derived exosomes containing LGTV grown in the presence of exosome-free FBS medium were also found to be infectious to HaCaT cells ( S3G Fig ) ., However , LGTV was not detectable in HaCaT cells ( grown in exosome-free FBS medium ) treated with the supernatant fractions ( S3G Fig ) ., Our data also showed that LGTV ( laboratory viral stocks , prepared from Vero cells ) was capable of infecting human vascular endothelial ( HUVEC ) cells with no differences in viral loads at 24 h p . i . in comparison to later tested time points ( 48 and 72 h p . i . ) ( S3H Fig ) ., HUVEC cells treated with exosomes-containing-LGTV showed significantly ( P<0 . 05 ) increased viral loads at 48 h p . i . in comparison to the cells treated with supernatant fractions , suggesting that LGTV RNA is enriched in exosomes ( S3I Fig ) ., We then performed transwell assays ( as described in the methods ) to test whether tick exosomes mediate transmission of LGTV from infected tick cells ( plated in upper inserts ) to uninfected/naïve human keratinocytes ( seeded into the lower well ) ., We found that tick cells treated with infected tick-cell-derived exosomes ( that were isolated from independent batch of LGTV-infected tick cells ) readily transmitted infectious exosomes to uninfected HaCaT cells ( Fig 1M ) ., However , upon incubations with tick cell-derived exosomes collected from GW4869 ( 5 μM ) treated cells , significantly ( P<0 . 05 ) reduced transmission of viral RNA to HaCaT cells was noted ( Fig 1M ) ., Infection of arthropod cells with laboratory virus stocks with known titers ( MOI 1 ) served as control in this assay ( Fig 1M ) ., Taken together , these results suggest that LGTV infectious RNA and proteins are transmitted to human cells via arthropod exosomes ., Upon transmission to the vertebrate host , arthropod-borne neurotropic encephalitis viruses are known to first replicate in the blood and peripheral tissues ( spleen and liver ) , cross the BBB and invade the CNS 58 , 59 ., We used mouse brain-microvascular endothelial cells ( bEnd . 3 cells; that constitutes the BBB ) to test whether LGTV infectious RNA and viral proteins are transmitted to neuronal cells via bEnd . 3 cell-derived exosomes ., LGTV readily infected and replicated in bEnd . 3 cells at all tested time points ( 48 and 72 h p . i . ) ( S4A Fig ) ., In addition , we found that the viral loads in brain endothelial cells were not significantly different over the infection period as revealed by the viral loads at much later time points ( 96 and 120 h p . i . ) ( S4B Fig ) ., QRT-PCR analysis revealed significantly ( P<0 . 05 ) increased viral burden and copy numbers in exosomes isolated from bEnd . 3 cells at 24 h p . i . in comparison to the other tested time points ( 48 , 72 , 96 and 120 h p . i . ) ( Fig 2A and S4C Fig ) ., We also detected higher loads of LGTV positive and negative sense RNA strands at 24 and 48 h p . i . , in comparison to the other tested time points ( 72 , 96 and 120 h p . i . ) ( Fig 2B ) ., LGTV infected and replicated in neuronal cells ( mouse N2a cells ) in a time-dependent manner with peak level of infection at 72 h p . i . ( S4D Fig ) ., N2a cells were then infected with bEnd . 3 cell-derived exosomes collected at different time points ( 24 and 48 h p . i . ) ., LGTV RNA and proteins containing exosomes from bEnd . 3 cells were found to be infectious to N2a cells with peak level of infection observed with exosomes isolated from endothelial cells at 48 h ( p . i . ) ( Fig 2C ) ., N2a cells treated with supernatant fractions ( collected at the indicated time points ) derived from endothelial cells resulted in significantly ( P<0 . 05 ) lower viral loads in comparison to the treatments with exosome fractions isolated from the bEnd . 3 cells ( Fig 2C ) ., Transwell assays performed with exosomes isolated from LGTV-infected-brain endothelial cells showed transmission of viral RNA and proteins from bEnd . 3 cells ( plated in upper inserts ) to uninfected/naïve N2a cells seeded in the lower well ( Fig 2D ) ., Presence of exosome inhibitor significantly reduced transmission of LGTV infectious RNA from bEnd . 3 cell-derived exosomes to N2a cells ( Fig 2D ) ., Infection of bEnd . 3 cells with laboratory virus stocks with known titers ( 6 MOI ) showed transmission of LGTV to N2a cells ( by crossing the membrane barriers in transwell plates ) and served as control in this assay ( Fig 2D ) ., These results suggest that exosomes derived from brain-endothelial cells are perhaps the mediators for BBB permeability ( crossing of infectious exosomes from infected-endothelial cells lining the BBB and transmission to the neuronal cells ) that may facilitate neuroinvasion of tick-borne LGTV and possibly TBEV and POWV ., Upon entry in to the brain , tick-borne neuroinvasive viruses ( such as TBEV ) infects neuronal cells 60 ., To test whether transmission of these viruses within the brain from one neuronal cell to another is mediated by exosomes; we first infected N2a cells with LGTV ( S4C Fig ) ., Cryo-EM showed the presence of purified exosome preparations from neuronal cell-derived exosomal fractions with the size range of 30 to 200 nm in diameter ( Fig 3A ) , similar to exosomes isolated from tick cells ., Also , we isolated exosomes by precipitation using the commercially available kit isolation reagent following the manufacturer’s protocol ( S5A Fig ) ., Cryo-EM images ( generated using this method ) showed the presence of purified exosome preparations from neuronal cell-derived exosomal fractions with the similar size range of 30 to 200 nm in diameter ( S5B Fig ) ., Like arthropod exosomes , neuronal cell-derived exosomes also showed a heterogenous population of vesicles ., In a very similar way , we did quantitative analysis using cryo-EM images collected from both uninfected and LGTV-infected N2a cell-derived exosomes ., Majority of these exosomes were also of sizes between 50–100 nm in both uninfected and infected groups ( Fig 3B and 3C ) ., Smaller exosomes of sizes 0–50 nm were of slight higher percentages in infected exosomes when compared to the uninfected group ( Fig 3B and 3C ) ., Fewer vesicles from sizes of 150–200 ( 9–11% ) or 200–250 ( 6 . 3% ) were found in both infected and uninfected groups ., Less than 1% of larger vesicles ( 250–350 nm sizes ) were found in infected group ( Fig 3B and 3C ) ., Counting of exosomes per image showed higher number of exosomes in LGTV-infected ( n = 13 ) in comparison to the uninfected ( n = 9 ) group ( Fig 3D ) ., This data suggested that LGTV-infection ( 72 h p . i . ) might enhance the production of exosomes ., The OptiPrep density gradient exosome separation ( that separates exosomes from viruses and large microvesicles ) yielded purified exosomes at six different fractions ., Immunoblotting analysis ( using 4G2 antibody ) of these fractions ( 20 μl ) showed presence of LGTV E-protein in all fractions but enriched amounts of E-protein were present in fractions four and five in comparison to the other fractions ( Fig 3E ) ., We did not detect E-protein in the fractions from uninfected control ., The cell lysates ( 20 μg ) from uninfected and infected groups were used as controls to compare the amounts of E-protein detected in the different fractions volume ( Fig 3E ) ., Immunoblotting with anti-HSP70 antibody detected enriched amounts of HSP70 ( exosomal marker ) in fourth fraction of both uninfected and infected samples ( Fig 3E ) ., HSP70 levels were also detected in three and five of infected fractions but not in uninfected fractions ( Fig 3E ) ., In addition to the HSP70 , we also analyzed the CD9 ( a protein enriched in the mammalian cell-derived exosomes and recognized as exosomal marker ) levels in uninfected and infected fractions ., CD9 was detected in all six of the uninfected fractions in an increasing manner , with higher levels in fractions four and five ( Fig 3E ) ., However , CD9 was detected in 2–5 of infected fractions with higher-level detection in fractions three and four ( Fig 3E ) ., It was interesting to note that LGTV E-protein was enhanced in similar exosomal fractions ( fractions 3–5 ) that had enhanced loads of both HSP70 and CD9 , suggesting that infectious exosomes in fraction four have higher levels of exosomal markers ., OptiPrep DG-isolation of exosomes using 0 . 1 μm filter ( culture supernatants were filtered before concentration and processing for gradient steps ) detected E-protein also in the fraction 4 , suggesting that these infectious exosomes have sizes of 50–100 nm ( Fig 3E ) ., This data also correlated with the quantitative analysis from cryo-EM images ., In order to address , where the intact LGTV particles may run on the parallel gradients , we performed OptiPrep DG-isolation on the laboratory stocks of LGTV ( prepared in Vero cells , collected at 7–14 days post-infection and stored at -80°C ) ., We noted a differential pattern in E-protein loads when density gradients were performed on LGTV-infected exosomal fractions from N2a cells ( Fig 3E ) or on LGTV laboratory stocks containing viruses ( S5C Fig ) ., An enhanced E-protein signal was detected in fraction 6 ( indicating presence of virions in this fraction ) and not in fraction 5 ., Detection of E-protein in fractions 4 , 3 and 2 from the laboratory virus stock suggested the presence of infectious exosomes containing viral E-protein ( S5C Fig ) ., This data indicated that the viral stocks are not just the virions but are perhaps mixtures of infectious exosomes containing viral E protein ., Upon LGTV infection of N2a cells , exosomes were collected at different time points ( 24 , 48 , 72 h p . i . ) and analyzed for viral loads ., QRT-PCR analysis revealed an increased total viral RNA load and copy numbers at 72 h p . i . in comparison to the other tested time points ( 24 and 48 h p . i . ) ( Fig 3F and 3G ) ., Both positive- and negative-sense RNA was detected at higher levels in the exosomes isolated from N2a cells at 72 h p . i . in comparison to the other tested time points ( Fig 3H ) ., Exosomes collected from the kit reagent also yielded similar results with increased LGTV loads in exosomes ( S5D Fig ) ., Next , we addressed the possibility that viral RNA could be binding to the outside of the exosomes and may be transmitted to the recipient cells ., In order to test this possibility , we treated freshly prepared LGTV-infected ( 72 h p . i . ) - N2a cell-derived exosomes with RNase A ( 5 μg/ml , for 15 min , at 37°C ) ., We did not find any differences in LGTV loads from infected treated or untreated groups ( Fig 3I ) ., The uninfected group treated with RNase A was kept as internal control ( Fig 3I ) ., In addition , we treated freshly derived exosomes isolated from LGTV-infected N2a cells , with Triton-X-100 ( 0 . 1%; for 45 min , at RT ) , followed by treatments with RNaseA ( 5 μg/ml , for 15 min , at 37°C ) ., QRT-PCR analysis showed that exosomes treated with both Triton-X-100 and RNaseA has lower LGTV loads in comparison to exosomes not treated with RNaseA ( S5F Fig ) ., Immunoblotting analysis further suggested the presence of LGTV E-protein in the exosomes isolated from N2a cells ( Fig 3J ) ., The E-protein loads were one-or two-fold higher in total lysates in comparison to the exosomal lysates derived from N2a cells ( Fig 3J ) ., Reduced molecular mass of LGTV E protein was found in exosomes derived from N2a cells in comparison to the total lysates ( Fig 3J ) , suggesting a possible de-glycosylation of the viral E protein in neuronal cell-derived exosomes ., We found similar de-glycosylation of the viral E protein in immunoblots performed on laboratory virus stocks ( S5E Fig ) ., A high level of CD9 was detected in the LGTV-infected N2a cell-derived exosomes in comparison to low levels in the uninfected control and the total cell lysates prepared from LGTV-infected or uninfected N2a whole cells ( Fig 3J ) ., Total protein lysates used in the immunoblot analysis served as loading control ( Fig 3J ) ., Enhanced levels of LGTV- E protein in neuronal exosomes treated with Triton-X-100 ( 0 . 03%; for 30 min , RT ) in comparison to the exosomes treated after freeze-thaw cycle ( thrice frozen and thawed at -80°C ) or untreated exosomes held at 4°C was detected by native-PAGE followed by immunoblotting with 4G2 antibody ( Fig 3K ) ., We noticed that E-protein was detected at higher molecular mass ( <250kDa ) in neuronal exosomes when samples were processed for native-PAGE analysis under non-reducing and non-denaturation conditions ., Detection of NS1 protein in independent samples at the similar molecular mass suggests presence of other LGTV proteins or polyprotein in exosomes ( Fig 3K ) ., Exosomes derived from uninfected N2a cells served as control ( Fig 3K ) ., Total protein lysates prepared from uninfected or LGTV-infected neuronal cell-derived exosomes after freeze-thaw or Triton-X-100 treatments or untreated samples served as loading control ( Fig 3K ) ., ELISA corroborate results of the native-PAGE , where higher loads of LGTV E-protein were detected when exosomes were treated with 0 . 1% of Triton-X-100 in comparison to untreated exosomal fractions ( Fig 3L ) ., Lower level of E-protein in LGTV-infected untreated neuronal exosomes was considered as background signal due to non-specific antibody binding ( Fig 3L ) ., Furthermore , we analyzed the presence of E-protein inside neuronal exosomes by a 4G2-antibody-coated bead-binding assay as described in methods ( Fig 3M ) ., No significant ( P>0 . 05 ) differences in viral loads were observed in LGTV-infected ( 72 h p . i . ) neuronal exosome samples that were untreated/treated with either 4G2 antibody or isotype control antibody ( Fig 3M ) ., GW4869 inhibitor treated exosomes from LGTV-infected neuronal cells collected at 72 h p . i . , followed by treatments with either 4G2 or isotype control also showed no significant ( P>0 . 05 ) differences in viral load in comparison to untreated samples ( Fig 3M ) ., However , a significant decrease in LGTV loads were observed in the inhibitor treated group in comparison to no-inhibitor treated group ( Fig 3M ) ., In addition , we found that exosomes treated with Proteinase K ( 100 μg/μl , 15 min at 37°C ) may be digested all proteins on the surface , thereby , lysing the vesicles and allowing degradation of the exosomal luminal proteins ( S5G Fig ) ., We detected E-protein in infected- untreated samples but not in treated samples ., Untreated , uninfected samples serve as internal controls ( S5G Fig ) ., The Ponceau S stained blot showed no proteins upon Proteinase K treatment ( S5G Fig ) ., Plaque assays further confirmed that exosomes isolated from LGTV-infected N2a cells contain infectious viral RNA , with a significantly higher number of plaques , evident upon infection with exosome fractions in comparison to the infection with supernatant fractions ( Fig 4A and S6A and S6B Fig ) ., Furthermore , infectious exosomes containing LGTV RNA and proteins prepared from N2a cells at different time points ( 24 , 48 , 72 h p . i . ) were capable of re-infecting naïve N2a cells ( Fig 4B ) ., Significantly higher level of viral burden was evident in N2a cells freshly infected with LGTV-containing exosome fractions prepared from 48 or 72 h ( p . i . ) in comparison to the infection with exosome fractions prepared from 24 h p . i . ( Fig 4B ) ., Re-infection with supernatant fractions showed undetectable levels of LGTV ( Fig 4B ) ., Similar levels of viral re-infection kinetics were observed upon incubations with LGTV-infected N2a cell-derived exosomes isolated using commercially available isolation reagent that were used to infect naïve/fresh N2a cells ( S6C Fig ) ., To find , if mosquito-borne flaviviruses such as WNV viral RNA is also present in exosomes , mouse N2a cells were infected with WNV ., Viral infection kinetics showed that WNV readily infected N2a cells with increased viremia at 72 h p

Introduction, Results, Discussion, Materials and methods

Molecular determinants and mechanisms of arthropod-borne flavivirus transmission to the vertebrate host are poorly understood ., In this study , we show for the first time that a cell line from medically important arthropods , such as ticks , secretes extracellular vesicles ( EVs ) including exosomes that mediate transmission of flavivirus RNA and proteins to the human cells ., Our study shows that tick-borne Langat virus ( LGTV ) , a model pathogen closely related to tick-borne encephalitis virus ( TBEV ) , profusely uses arthropod exosomes for transmission of viral RNA and proteins to the human- skin keratinocytes and blood endothelial cells ., Cryo-electron microscopy showed the presence of purified arthropod/neuronal exosomes with the size range of 30 to 200 nm in diameter ., Both positive and negative strands of LGTV RNA and viral envelope-protein were detected inside exosomes derived from arthropod , murine and human cells ., Detection of Nonstructural 1 ( NS1 ) protein in arthropod and neuronal exosomes further suggested that exosomes contain viral proteins ., Viral RNA and proteins in exosomes derived from tick and mammalian cells were secured , highly infectious and replicative in all tested evaluations ., Treatment with GW4869 , a selective inhibitor that blocks exosome release affected LGTV loads in both arthropod and mammalian cell-derived exosomes ., Transwell-migration assays showed that exosomes derived from infected-brain-microvascular endothelial cells ( that constitute the blood-brain barrier ) facilitated LGTV RNA and protein transmission , crossing of the barriers and infection of neuronal cells ., Neuronal infection showed abundant loads of both tick-borne LGTV and mosquito-borne West Nile virus RNA in exosomes ., Our data also suggest that exosome-mediated LGTV viral transmission is clathrin-dependent ., Collectively , our results suggest that flaviviruses uses arthropod-derived exosomes as a novel means for viral RNA and protein transmission from the vector , and the vertebrate exosomes for dissemination within the host that may subsequently allow neuroinvasion and neuropathogenesis .

In this study we have demonstrated that cells from the medically important vector tick , secretes exosomes that mediate transmission of tick-borne Langat ( LGTV ) viruses from arthropod to human and other vertebrate host cells ., This study not only provides evidence that suggest tick-borne pathogens use arthropod-derived exosomes for transmission from vector to mammalian cells but also use exosomes for dissemination within the vertebrate host .

invertebrates, medicine and health sciences, vesicles, viral transmission and infection, molecular probe techniques, endothelial cells, rna extraction, immunoblotting, microbiology, vertebrates, neuroscience, animals, epithelial cells, molecular biology techniques, cellular structures and organelles, extraction techniques, viral load, research and analysis methods, animal cells, biological tissue, molecular biology, arthropoda, cellular neuroscience, eukaryota, cell biology, anatomy, exosomes, virology, neurons, epithelium, biology and life sciences, cellular types, organisms

journal.pntd.0005813

A dynamic model for estimating adult female mortality from ovarian dissection data for the tsetse fly Glossina pallidipes Austen sampled in Zimbabwe

Human and animal trypanosomiasis , which is spread by tsetse flies ( Glossina spp ) , is a major health concern in much of sub-Saharan Africa 1–3 ., Research on trypanosomiasis and tsetse has been carried out for nearly 60 years at Rekomitjie Research Station in the Zambezi Valley ( 16° 18 S , 29° 23 E; altitude 500 m ) 4 ., Studies at the station have provided improved understanding of vector and disease dynamics , with the aim of improving disease control 5–7 ., It has been shown that the basic reproduction number of vector borne diseases such as trypanosomiasis is strongly dependent on vector mortality rates 8 , 9 ., Accurate estimates of adult tsetse mortality constitute an essential element of that understanding 10 ., Laboratory animals and wild populations can differ greatly in their life expectancies 11 ., To understand population dynamics in the wild , it is thus essential to obtain data from free-ranging field populations rather than laboratory animals 11 ., Mark-recapture can provide good estimates of tsetse population parameters in closed situations , particularly where it is feasible to recapture the same flies many times during their lifetimes 12 , 13 ., Mark-recapture studies are , however , logistically demanding , costly , and time consuming 10 ., Moreover , in open populations subject to in- and out- migration , the results are often difficult to interpret , and researchers have accordingly developed alternate methods for estimating mortality 10 , 14 ., As described more extensively elsewhere 15 , 16 , ovarian dissection of tsetse can show the number of times a fly has ovulated ., Since tsetse ovulate at approximately regular intervals , these data can be used to estimate the age distribution of female tsetse populations ., It has been argued that it should then be possible , in principle , to determine how female mortality rates change with season , and over time , by analysing changes in age-distributions of female tsetse 17–20 ., Current techniques for estimating female mortality from ovarian dissection data rely on three important assumptions 17: first , that sampling probability is not dependent on the age of the fly; second , that mortality rates are independent of the age of the fly; third , and crucially , that the population under study has a stable age distribution ., A recent study shows , however , that these assumptions are often violated , leading to unrealistic mortality estimates ., For example , standard techniques predict that mortality decreases with increasing temperature , which contradicts data from mark-recapture studies and is unlikely on biological grounds 10 ., We therefore need new methods , for estimating female mortality from ovarian dissection data , which allow for unstable age distributions and for age-related changes in mortality and capture probability ., We develop dynamic models that simulate female tsetse populations , and the associated changes in their age distribution ., At Rekomitjie , instability of the age distribution appears to result from large seasonal variation in temperature ., Elsewhere , such instability could , however , result from other factors , such as seasonal changes in host density or rainfall ., Our alternative approach to mortality estimation will be appropriate however the instabilities arise ., Our models estimate or incorporate published estimates of temperature-dependent mortality in adults and pupae , temperature-dependent development rates in pupae , and density-dependent mortality in pupae 16 , 21 , 22 ., The present paper is concerned with adult female G . pallidipes captured between 1 July 1991 and 30 June 1992 at Rekomitjie using stationary mechanical traps 23 , baited with artificial host odour , consisting of acetone ( dispensed at 500 mg/h ) , 1-octen-3-ol ( 0 . 4 mg/h ) , 4-methylphenol ( 0 . 8 mg/h ) and 3-n-propylphenol ( 0 . 1 mg/h ) 24 ., The capture and processing methods have been described in detail elsewhere 5 ., While fly populations and capture probabilities vary seasonally , trap catches were approximately the same at the start and end of the study period ., Female tsetse flies were subjected to ovarian dissection , and assigned to an ovarian category , depending on the relative sizes of the oocytes in the left and right ovaries 15 , 16 ., For flies that have ovulated fewer than four times the ovarian category is equal to the number of times that the fly has ovulated ., For flies that have ovulated more than three times , the ovarian category provides only the number of ovulations , or ovarian age , modulo four: that is , a fly in ovarian category 4 may have ovulated 4 , 8 , 12 , 16 … etc . times , and analogous statements apply to flies in ovarian categories 5 , 6 and 7 ., Flies too damaged to assign an ovarian category were excluded from the current analysis ., During the study , 19 , 323 female G . pallidipes were dissected and assigned an ovarian category ., The data were aggregated into monthly counts of flies in each category , as detailed in the Supporting Information ., Daily maximum and minimum temperatures are routinely measured using mercury thermometers housed in a Stevenson screen at the station ., We incorporated smoothed monthly mean temperatures ( estimated as the monthly average of the maximum and minimum readings for each day of the month ) in our model of female tsetse populations ., Daily mean temperatures are given in the Supporting Information ., We develop a deterministic compartmental model of female tsetse populations based on an understanding of tsetse biology acquired from field and laboratory data ., We model only the female population since this is the productive part of the population , for which we have age distribution data ., Male and female G . pallidipes at Rekomitjie accrue mouthpart and mid-gut trypanosome infections at similar rates but , since females also live significantly longer than males , they are more likely to transmit trypanosomes ( 3 ) ., There are always sufficient G . pallidipes males at Rekomitjie such that about 98% of females are inseminated by the age of 8 days 16 ., We do not , therefore , need to model the male population in order to model the dynamics of the female population ., Our model assumes that all females are inseminated and ovulate at 6 days in age ., The schematic of the model is shown in Fig 1 and parameters are described in Table 1 ., Tsetse could be in one of nine groups: the pupal stage and adult ovarian ages 0 to 7 ., The assumed pattern of mortality among adult female tsetse is based on field experiments from Zimbabwe 17 , 18 , 21 ., The experiments showed that adult female G . m ., morsitans suffered mortality in excess of 10% per day immediately after emergence: mortality then declined rapidly to about 3% by age 8 days , and then to about 1% per day at age 20 days ., Thereafter , mortality increased steadily but slowly with age , such that the mortality was still less than 2 . 5% per day at age 100 days ., Female adult mortality is thus clearly a function of age , but the major changes are restricted to the immature stages , while the fly’s thoracic musculature is still developing , and before it has ovulated for the first time ., Notice that , since we are fitting our models only to ovarian age data from adult females , and we have no data on pupal numbers or losses , we cannot separate between deaths that occur in the pupal stage proper and those that occur among flies that have just emerged from the pupa and are not yet available to traps ., Thus , while for convenience , we formally apply temperature and density-dependent mortality to pupae in the modelling , pupal mortality needs , strictly , to be interpreted as mortality in all stages prior to the mature adult stage ., In subsequent text and tables , we refer to this mortality as immature mortality ., We assumed , as a first approximation , that mortality was constant for all mature adult females ., We also include , however , a model in which mature adult female mortality was allowed to vary with age in the Supporting Information ., Adults in a given age category are assumed to progress to the next age category at a constant rate , with flies of ovarian age 0 progressing to ovarian age 1 at a rate of ( 6 days ) -1 and flies of all other ages progressing to the subsequent ovarian ages at a rate of ( 9 days ) -16 , 7 , 16 , 25 ., These rates are all temperature dependent but , at the field temperatures observed at Rekomitjie , the times between ovulations are unlikely to vary by more than 2 days from the assumed mean value and should not thus be a source of major error in the mortality estimates ., Pupae are deposited at a rate equal to the number of flies in ovarian categories 1 to 7 divided by the expected amount of time ( 9 days ) which flies remain at these ovarian ages ., Since only half of these pupae are female , this quantity is then divided by two to obtain the number of female pupae deposited ., In other settings a pupal mortality of around 1% per day 26 has been assumed; in order to encompass this estimate in our models , we set the minimum pupal mortality rate below this at 0 . 001 per day ., There is evidence for density-dependent mortality in pupae in the wild 26: accordingly , we model a density-dependent pupal mortality rate proportional to the number of female pupae ., The pupal mortality rate at temperatures under 25°C is given by:, μp , T≤25=0 . 001+dP ,, ( 1 ), Where d parameterizes the density dependence and P is the pupal population in a given , but undefined , area ., Pupal duration is temperature-dependent in tsetse and female pupa emerge to become adults in age category zero ., Laboratory studies show that pupal emergence rates and temperature are related by 27:, h ( T ) =k/ ( 1+exp ( a+bT ) ), ( 2 ), where T is the mean daily temperature and k , a , and b are constrained to be the values given in Table 1 ., We assume that Eq ( 2 ) also provides acceptable estimates of pupal duration for female G . pallidipes at Rekomitjie ., Mark-recapture studies suggest that mortality in adult female G . pallidipes increases exponentially with temperature ( T ) for T > 25°C 28 ., Accordingly , we model adult fly mortality using:, μa ( T , T>25°C ) =μa , T≤25expβ ( T−25 ), ( 3 ), Where μa , T≤25 is the mature adult mortality for temperatures at or below 25°C , and β parameterizes the increase at higher temperatures ., Pupal mortality is also thought to be temperature dependent and we also allow this quantity to increase exponentially with temperatures above 25°C:, μp ( T , T>25°C ) =μp , T≤25expα ( T−25 ) ,, ( 4 ), Where μp , T≤25 is the adult mortality rate for temperatures at or below 25°C given in Eq 1 , and α parameterizes the increase in the mortality rate at higher temperatures ., Field evidence from Zimbabwe indicates the probability with which a female tsetse is captured in a trap increases with her age , with the greatest bias against flies in ovarian categories 0 and 1 5 , 29 ., We model this by allowing a relative risk of capture less than 1 for these young flies ., All simulation and analysis was performed using R 3 . 3 . 0 ., Differential equations specifying the three models , based on the compartmental model diagram in Fig 1 , are given in the Supporting Information ., Differential equations were solved numerically using the Livermore solver in the deSolve package ., Since starting conditions were unknown , models were run for three years , which allows enough time for a stable pattern in population levels to be achieved ., For the last year of the simulation , the average ovarian age distribution was calculated for each month of the study and compared with the observed ovarian age distribution for that month ., For the main analyses , the log-likelihood for a given month was calculated assuming multinomial sampling of the available population ., For the fits to both ovarian dissection and population data , the log-likelihood of the population data given the model was calculated using binomial sampling probabilities and added to the log-likelihood of the ovarian dissection data ., To do this , catches per trap were scaled such that equal weight was given to both datasets , the true population was scaled such that it was large compared to the catch population ( 10 , 000 times on average ) , and the sampling probability was assigned to be the mean ratio of the scaled catch population to the true population ., The overall log-likelihood was calculated as the sum of the log-likelihoods for each month ., Since the population was roughly the same at the beginning and the end of the study period , a simulation was given a likelihood of zero if the simulation failed to produce a pupal population on July 1 , 1991 within 3% of the pupal population on June 30 , 1992 ., Rates of pupal development , of transition between ovarian categories , and of minimum pupal mortality were fixed at the values suggested in the literature: all other parameters were determined using model fits ., For each model , the maximum log-likelihood of the model was determined using the Nelder-Mead downhill-simplex algorithm , as implemented using the optim function in R . The optimization was performed to determine the optimum parameter set for that model , using a logit scale for some parameters ( S0 , S1 , μa ) and a log scale for the remainder of the unknown parameters ., The Hessian was calculated and then inverted to give the Fisher information matrix , which was used to obtain confidence limits for the variables ., The confidence limits were then detransformed to obtain the confidence intervals for each parameter ., Nested models that have an AIC at least 6 units larger than the best performing model were considered to confer a significantly worse fit than the best performing model ( less than 0 . 05 times as probable as the best performing model ) 30 ., Standard techniques for estimating female tsetse mortality from ovarian dissection data are biased , especially during the hot-dry season ( November to March ) 10 ., The bias associated with standard techniques varies with temperature variation: where , as in Zimbabwe , this variation is large , the bias will be most severe ., Tsetse near the equator , where temperature variations are small , may have age distributions that are approximately stable—although it is possible other environmental factors may lead to nonstable age distributions in these settings 10 , 25 ., To quantify this bias , using simulated data from our model together with our the maximum likelihood parameters , we estimated daily mortalities for the study period using standard techniques , as described elsewhere 15 , 17 , 26 , 29 and reviewed by Hargrove and Ackley 10 ., Using maximum likelihood parameter estimates , we generated simulated female tsetse populations with varying magnitudes of temperature variation , but the same mean temperature ., For a parameter f , varying between 0 and 1 , we created counterfactual daily temperature Tnew ( f ) for each day based on the true temperature in Zimbabwe , T , as follows:, Tnew ( f ) =f ( T−T¯ ) +T¯ ,, ( 5 ), Where T¯ is the mean annual temperature ., We then determined the mean bias in mortality estimation using standard techniques as a function of f ., For the main analysis , we fit models 1–3 to the age distribution data , only constraining the population to be approximately the same at the beginning and end of the experimental period , but not taking any account of differences between predicted population levels and observed trap catches in the period between the endpoints ., Maximum likelihood fits for model 3 provide a good fit to the age distribution data , but show large differences between predicted trends in simulated population and trap catches ( Fig 2 ) ., Maximum likelihood fits are shown for models 1 and 2 in the Supporting Information ., Table 2 summarizes the estimates of maximum likelihood parameters and their confidence intervals ., Model 3 , which allows for differential mortality between mature adults and immature tsetse , confers a significantly better fit than models 1 and 2 ., For all three models , the estimated mortality rate at temperatures less than 25°C is consistently 0 . 03 per day; 3% attrition per day is considered plausible for a non-decreasing tsetse population 12 , 31 , and the maximum adult mortality before the population starts to decline is estimated to be approximately 4% per day 32 ., For model 3 , we estimate a mortality rate of 0 . 028 per day at temperatures less than 25°C , with mortality increasing for temperatures above 25°C ., Figs 3 and 4 illustrate that mortality among mature adults hardly changes with time of year and temperature , whereas there is significantly more variation in the mortality of immature tsetse , which exhibit peaks in about November and March ., Pupal mortality exceeds mature adult mortality except during the dry , generally cooler , months of May to September ., When model 3 was fit to both the ovarian dissection data and the catch data simultaneously there was a poorer fit to the ovarian dissection data , as apparent in the considerably lower likelihood and from examination of the fits ( Table 3 , Fig 5 ) ., Using this model , it is not possible to simultaneously obtain good fits to both the age distribution and the trap catch data ., Nonetheless , fitting to both data sources produces a mortality rate estimate for mature adults of 0 . 027 per day at temperatures under 25°C , comparable to estimates from the models fit only to the ovarian dissection data ( Table 2 ) ., Given the challenges in fitting both data sources and the uncertainty of the correspondence between trap catches and population levels , we used model fits to only the age distribution data for further investigations ., We used model 3 , the minimum AIC model in our main analysis , and corresponding parameter estimates ( Table 2 ) to quantify the possible biases inherent in standard mortality estimation techniques ., We generated daily ovarian category distributions using observed smoothed monthly mean temperatures at Rekomitjie ., With these distributions , we estimate the mortality among mature adult tsetse for that day ., Fig 6 shows the model-generated mortality ( solid line ) and the mortality estimated using standard techniques ( dashed line ) as a function of time of year at Rekomitjie , assuming the maximum likelihood parameter estimates for model 3 ( Table 2 ) ., Standard techniques are biased at all times of year , but the bias is largest during the hot-dry season ., In this season , standard techniques give the lowest mortality estimates when mortality is in fact highest ., Our modelling suggests , however , that this increased mortality occurs almost exclusively among recently emerged , immature , adults ., Since data are typically aggregated by month to estimate a monthly mortality , Fig 6 also shows the monthly mean of mortalities generated using our model ( closed dot ) and estimated using standard techniques ( open dot ) ., Aggregating the data by month does not significantly alter the magnitude of the bias ., Fig 7 shows the mean bias of standard mortality estimation techniques as a function of temperature variation ., This bias is plotted for various mortalities ( text annotations ) , using model 3 and the maximum likelihood parameter estimates for the remaining parameters ., As shown in Eq 5 , f = 0 corresponds to no temperature variation , and f = 1 corresponds to the temperature variation at Rekomitjie ., The general trend is that bias in standard techniques is greater when there is greater temperature variation ., However , the magnitude of the bias and the extent to which it depends on temperature variation depends on the mortality rate ., A major finding of this study is that mortality in immature tsetse is higher , and increases much more rapidly with increasing temperature , than in mature adults ., These results are consistent with results from a mark-recapture study showing that mortality among recently emerged female tsetse is markedly higher than for all older flies 16 , 21 ., It may be objected , however , that the high mortality estimated in the mark recapture experiment , in newly emerged—and newly marked and released—flies might merely reflect stress due to their handling and marking ., This possibility was acknowledged in the original analysis 33 , but it was argued that the continual decrease in the loss rate over the first 18 days of life was consistent with high ( natural ) losses in young flies ., This conclusion is also consistent with published evidence that a large percentage of newly emerged tsetse can die before they become available for capture in the field and that this percentage can increase dramatically at high temperatures 31 , 33–35 ., The results are , admittedly , at variance with published data on other insects where there have been no reports of mortality being higher in recently emerging insects than in older adults ., In part , however , this may reflect the difficulties attendant on estimating insect mortality in the field ., The case of age-dependent mortality in mosquitoes provides a good example of the problems involved ., The following analysis of published methods for estimating age-specific mortality in mosquitoes suggests that they would not allow the detection of increased mortality in newly emerged field mosquitoes even if it existed ., To our knowledge , nobody has yet carried out on mosquitoes , or indeed on any other insect , the equivalent of the experiment on tsetse where insects were marked uniquely , released in field at birth , and where their recapture history was recorded for the rest of their lives 16 , 21 ., Early analysis of the age structure of mosquitoes captured in the field showed that mortality rates increased with age 36 ., There was no evidence for increased mortality in the youngest mosquitoes , but this method obviously cannot estimate the number of mosquitoes that die before they become available for sampling and thus cannot provide any estimate of mortality among the youngest mosquitoes ., This objection does not apply to a study where the survival of large samples ( >10 , 000 ) of mosquitoes was followed from birth 37 ., For both sexes , mortality was low at young ages ( < 10 days old ) , steadily increased among middle-aged mosquitoes , and decelerated at older ages ., Again , therefore , this experiment produced no evidence for increased mortality among young mosquitoes ., However , this was a study of laboratory-bred and raised mosquitoes and , again , says nothing about the rate at which newly emerged mosquitoes might die in the field ., As observed above and elsewhere , mortality among young tsetse in the laboratory is much lower than estimated for field flies 21: the same may well be true for mosquitoes ., The “captive cohort method” for estimating population age structure in the wild can be used to estimate age-specific mortality rates ., Since , however , the method involves following the survival of samples captured in the field 38–40 , the problem referred to above arises: the method cannot provide mortality estimates for insects that die before they can be sampled ., Moreover , while the method is valid for stationary populations ( stable age distribution and zero growth rate ) , violations of this assumption require more complex modelling approaches , with quantification of population birth rates or immigration/emigration rates 39 ., Thus , while the studies reviewed here produced no evidence for increased mortality in very young mosquitoes , the methods used would not anyway be able to detect increased mortality among newly emerged mosquitoes in the field ., Nonetheless , it is not unreasonable to expect that the risks faced by young tsetse , relative to mature adults , are much greater than those faced by their mosquito counterparts ., Teneral tsetse have low levels of fat , poorly developed flight musculature , relatively weak flight capability and , being obligate blood feeders , need to locate a vertebrate host and feed off it safely before they starve 41–43 ., The problems for newly emerged mosquitoes are less severe: the flight performance of Aedes aegypti , for example , is highest during the first 14 days of life 44: since , also , mosquitoes can feed off nectar and plant juices , their feeding risks should be much lower than for tsetse ., Since we were only fitting our models to age distributions of adult females , we had no way of separating death rates in immature tsetse between deaths that occurred in the pupal stage and those occurring in very young adults , before they became available for trap capture ., Regardless of how these deaths among immature classes are counted , however , if they occur in large numbers they will contribute to destabilisation of the population age structure ., This effect has bedevilled past efforts to estimate female tsetse mortality from ovarian dissection data , which assumed that the population age distribution was stable 10 ., We developed dynamic models that allow for age distributions that are unstable , in our case due to fluctuations in temperature , and also allow for age-related changes in mortality and capture probability ., For each of the three models , mortality estimates are around 0 . 03 per day for temperatures under 25°C , consistent with estimates from other areas ., Mark-recapture studies at the nearby Antelope Island gave an adult female mortality rate of 0 . 023 at lower temperatures , but with a larger increase in mortality with increasing temperature than predicted by model 3 33 ., Previous mark-recapture work has shown that the increase in mortality for adult G . pallidipes females at higher temperatures is exponential and characterized by a coefficient of 0 . 106 33 , which is much larger than the value of the analogous parameter ( β ) estimated for model 3 . The disagreement could be due to the incorporation in the mark-recapture estimates of some young flies that still have higher natural mortality rates than fully mature flies ., Mortality is highest among flies that have just emerged , but only declines to mature levels over the first 10 days of adult life 33 ., Our modelling produces mortality estimates that appear more reliable than those from classical analyses in that mortality is predicted to increase with temperature , as expected ., Our technique also offers insight into the factors driving the dramatic changes in age distribution observed during the hot-dry season ., Model 2 , where mortality among mature adults and all immature tsetse have the same dependence on temperature , does not perform significantly better than model 1 , where mortality is independent of temperature ., Temperature-dependent mortality for immature stages and mature adults cannot thus explain the observed changes in age distribution if it is applied evenly across all stages ., Model 3 , which allows for differential dependence of mortality on temperature for mature adults and immature stages , performs significantly better than models 1 and 2 . This differential dependence of mortality on temperature causes the observed ratio of young flies to older flies to decrease at high temperatures , which would explain why a greater fraction of flies sampled during the hot-dry season are older than at other times of year ., Our results suggest that increases in temperature affect the mortality of recently emerged adult G . pallidipes more than all older flies ., Age-dependent mortality has been documented in the field for G . m ., morsitans 21 and in the laboratory for various tsetse species 22 , 45 , 46 , and we suggest that these differences may not remain constant with changes in temperature ., Using standard mortality estimation techniques to estimate fly mortality can demonstrate the bias inherent in these techniques and show how this bias varies seasonally for locations near and far from the equator ., We find that during the hot-dry season , the mortality estimates using standard techniques are most biased , whereas during the rest of the year , the bias is smaller , though still significant ( Fig 5 ) ., We also find that the bias in standard techniques will tend to decrease with decreasing annual variation in temperature ., Standard techniques , which assume a population that is declining or growing exponentially 17 , 29 , may nonetheless give biased mortality estimates , and the magnitude of this bias is not easily predicable since it may depend on parameters such as the mature adult mortality ., An unpredictable bias can make it impossible to compare mortalities at different times or in different regions ., Elsewhere in Africa , particularly at sites close to the Equator , it may be true that temperature variations are not sufficient to produce age structure instability ., However , the possibility cannot be excluded that age structure instability could result elsewhere from other climatological effects: any such effects will cause errors in mortality estimates where these are derived using the classical approach ., It is thus incumbent on investigators to convince themselves whether or not their study population does indeed exhibit a stable age structure ., Our work has several strengths:, ( i ) Extension of traditional mortality estimation techniques from ovarian dissection data ., Our model is an extension of traditional mortality estimation techniques from ovarian dissection data that allows for non-stable age distributions and explicitly constrains population growth ., This allows us to make mortality estimates that are directly comparable to those from traditional techniques , while avoiding incorrect assumptions ., ( ii ) Dynamical modelling ., Dynamical modelling techniques , such as the compartmental modelling we employ , have been used to estimate key demographic and disease transmission parameters in many settings , including for vector populations 47 and tsetse populations 12 ., ( iii ) Model based on an understanding of tsetse biology acquired from field and laboratory data ., Our models include temperature-dependent pupal emergence , pupal ( and immature ) mortality , and mature adult mortality , as well as density-dependent pupal mortality ., Our work also has several limitations ., ( i ) Unknown populations of adults and pupae over the course of the year ., The most serious challenge we have in our modelling arises from our ignorance regarding the way in which the true population numbers of adult and puparial tsetse vary with time and season ., Differences in tsetse trap catches from month to month are undoubtedly related to population changes; however , they also plausibly reflect changes in fly behaviour , and thus capture probability , with changing temperature , and potentially reflect changes in age structure and seasonal in- and out-migration 48 ., For our main analysis , we did not constrain our models to fit population numbers , whether for adult or immature stages ., The different models presented in Tables 2 and 3 give very different estimates for how the pupal and adult fly populations change over the course of the year ., Nonetheless , the mortality estimates for temperatures under 25°C did not vary significantly between the three models , and , for model 3 , with or without fitting to trap data ., The mortality estimates for temperatures less than 25°C do not , therefore , appear to be particularly sensitive to the way in which the total population is predicted to change ., However , more accurate estimates of population changes over the course of the year would be required to determine how high temperatures affect mortality ., ( ii ) Classification of teneral deaths ., As detailed in the Methods section , we could not separate pupal deaths from those occurring in newly emerged adults before they could be trapped ., Indeed , it is not always clear how to categorise deaths among immature classes ., Pupae that are predated or parasitized are clearly true pupal deaths ., A major loss of immature flies is , however , d

Introduction, Methods, Results, Discussion

Human and animal trypanosomiasis , spread by tsetse flies ( Glossina spp ) , is a major public health concern in much of sub-Saharan Africa ., The basic reproduction number of vector-borne diseases , such as trypanosomiasis , is a function of vector mortality rate ., Robust methods for estimating tsetse mortality are thus of interest for understanding population and disease dynamics and for optimal control ., Existing methods for estimating mortality in adult tsetse , from ovarian dissection data , often use invalid assumptions of the existence of a stable age distribution , and age-invariant mortality and capture probability ., We develop a dynamic model to estimate tsetse mortality from ovarian dissection data in populations where the age distribution is not necessarily stable ., The models correspond to several hypotheses about how temperature affects mortality: no temperature dependence ( model 1 ) , identical temperature dependence for mature adults and immature stages , i . e . , pupae and newly emerged adults ( model 2 ) , and differential temperature dependence for mature adults and immature stages ( model 3 ) ., We fit our models to ovarian dissection data for G . pallidipes collected at Rekomitjie Research Station in the Zambezi Valley in Zimbabwe ., We compare model fits to determine the most probable model , given the data , by calculating the Akaike Information Criterion ( AIC ) for each model ., The model that allows for a differential dependence of temperature on mortality for immature stages and mature adults ( model 3 ) performs significantly better than models 1 and 2 ., All models produce mortality estimates , for mature adults , of approximately 3% per day for mean daily temperatures below 25°C , consistent with those of mark-recapture studies performed in other settings ., For temperatures greater than 25°C , mortality among immature classes of tsetse increases substantially , whereas mortality remains roughly constant for mature adults ., As a sensitivity analysis , model 3 was simultaneously fit to both the ovarian dissection and trap data; while this fit also produces comparable mortality at temperatures below 25°C , it is not possible to obtain good fits to both data sources simultaneously , highlighting the uncertain correspondence between trap catches and population levels and/or the need for further improvements to our model ., The modelling approach employed here could be applied to any substantial time series of age distribution data .

Trypanosomiasis , spread by tsetse flies ( Glossina spp . ) , is a disease that is fatal for both humans and livestock if left untreated , and is a serious threat to public health in many regions of sub-Saharan Africa ., In order to understand the dynamics of the disease it is important also to understand tsetse population dynamics ., Tsetse fly mortality estimates are central to this understanding , but are difficult to acquire from wild populations ., Previous methods for estimating mortality from age-distribution data assume a stable age structure and age-invariant mortality and capture probability ., Based on prior fieldwork , none of these assumptions appears justified ., Building on previous mortality estimation techniques , and incorporating what is known about tsetse population dynamics , we develop simulation techniques to estimate mortality for tsetse populations where the age distribution is not necessarily stable ., We fit our models to age-distribution data produced in 1991 and 1992 at Rekomitjie Research Station in the Zambezi Valley in Zimbabwe ., Our final model produces mortality estimates consistent with those of mark-recapture studies performed in other settings ., We find that mortality increases with temperature , a result consistent with field and laboratory findings , and that the temperature effects are much more severe for pupae and newly emerged adults than for mature adults ., Our dynamical modelling approach could be used for mortality estimation for any population where substantial age distribution data are available: specifically , it could be used to answer substantive questions about tsetse flies in other settings .

death rates, invertebrates, medicine and health sciences, demography, age distribution, parasitic diseases, animals, glossina, developmental biology, pupae, tsetse fly, insect vectors, infectious diseases, zoonoses, epidemiology, life cycles, protozoan infections, trypanosomiasis, disease vectors, insects, arthropoda, people and places, mosquitoes, disease dynamics, biology and life sciences, species interactions, organisms

journal.pgen.1001289

A New Testing Strategy to Identify Rare Variants with Either Risk or Protective Effect on Disease

Common diseases such as diabetes , heart disease , schizophrenia , etc . , are likely caused by a complex interplay among many genes and environmental factors ., At any single disease locus allelic heterogeneity is expected , i . e . , there may be multiple , different susceptibility mutations at the locus conferring risk in different individuals 1 ., Common and rare variants could both be important contributors to disease risk ., Thus far , in a first attempt to find disease susceptibility loci , most research has focused on the discovery of common susceptibility variants ., This effort has been helped by the widespread availability of genome-wide arrays providing almost complete genomic coverage for common variants ., The genome-wide association studies performed so far have led to the discovery of many common variants reproducibly associated with various complex traits , showing that common variants can indeed affect risk to common diseases 2 , 3 ., However , the estimated effect sizes for these variants are small ( most odds ratios are below ) , with only a small fraction of trait heritability explained by these variants 4 ., For example , at least loci have been identified for height , but these loci together explain only of the estimated heritability for this trait 5 ., One possible explanation for this missing heritability is that , in addition to common variants , rare variants are also important ., Evidence to support a potential role for rare variants in complex traits comes from both empirical and theoretical studies ., There is an increasing number of recent studies on obesity , autism , schizophrenia , epilepsy , hypertension , HDL cholesterol , some cancers , Type-1 diabetes etc ., 6–15 that implicate rare variants ( both single position variants and structural variants ) in these traits ., From a theoretical point of view , population genetics theory predicts that most disease loci do not have susceptibility alleles at intermediate frequencies 16 , 17 ., With rapid advances in next-generation sequencing technologies it is becoming increasingly feasible to efficiently sequence large number of individuals genome-wide , allowing for the first time a systematic assessment of the role rare variants may play in influencing risk to complex diseases 18–21 ., The analysis of the resulting rare genetic variation poses many statistical challenges ., Due to the low frequencies of rare disease variants ( as low as , and maybe lower ) and the large number of rare variants in the genome , studies with realistic sample sizes will have low power to detect such loci one at a time , the way we have done in order to find common susceptibility variants 5 , 22 ., It is then necessary to perform an overall test for all rare variants in a gene or , more generally a candidate region , under the expectation that cases with disease are different with respect to rare variants compared with control individuals ., Several methods along these lines have already been proposed ., One of the first statistical methods proposed for the analysis of rare variants 23 is based on testing whether the proportion of carriers of rare variants is significantly different between cases and controls ., A subsequent paper by Madsen and Browning 24 introduced the concept of weighting variants according to their estimated frequencies in controls , so that less frequent variants are given higher weight compared with more common variants ., Price et al . 25 extended the weighted-sum approach in 24 to weight variants according to externally-defined weights , such as the probability of a variant to be functional ., One potential drawback for these methods is that they are very sensitive to the presence of protective and risk variants ., We introduce here a new testing strategy , which we call replication-based strategy , and which is based on a weighted-sum statistic , but that has the advantage of being less sensitive to the presence of both risk and protective variants in a genetic region of interest ., We illustrate the proposed approach on simulated data , and a real sequence dataset on Type-1 diabetes ., If external information is available on the plausibility of a rare variant to be related to disease , it is of interest to be able to incorporate such information into our testing strategy ., Such information has proved essential in the mapping of the disease genes for two monogenic disorders 26 , and may well prove important for mapping disease genes in more complex diseases ., It is straightforward to extend the proposed approach to take into account such information ., If we denote by the probability that a variant is functional , then we can rewrite the statistic above as:where signifies that variant occurs times in controls , and times in cases ., In particular , if for all variants then we recover the statistic S above , where functional information was not used ., If on the other hand a variant is not functional , then , and this variant is ignored ., We also applied our approach to a dataset on Type 1 Diabetes ( T1D ) , published by Nejentsev et al . 15 ., In their paper , the authors resequenced exons and splice sites of ten candidate genes in cases and controls ( more details on the dataset are in Text S2 ) ., In their study , rare variants were tested individually , and two SNVs in gene IFIH1 and two other SNVs in gene CLEC16A were found to be protective against T1D ., Here we reanalyze the dataset using the proposed approach , and two of the existing approaches ., For each gene and each method , we perform two-sided tests , testing for the presence of risk or protective variants ., Results are in Table 5 ., As in 15 we found one gene , IFIH1 , to be significant with all three methods ( P-value for all three methods ) ., For this gene , controls were enriched for rare mutations compared with cases ., Some evidence of enrichment in protective variants was also observed in another gene , CLEC16A , although the P-values do not remain significant after multiple testing correction ., We have proposed here a new testing strategy to examine associations between rare variants and complex traits ., The approach is based on a weighted-sum statistic that makes efficient use of the information the data provides on the presence of disease variants in the region being investigated ., The proposed test is based on computing two one-sided statistics , designed to quantify enrichment in risk variants , and protective variants , respectively ., This aspect allows the proposed approach to have substantially better power than existing approaches in the presence of both risk and protective variants in a region ., Even when only one kind of variants is present , we have shown via simulations that the proposed approach has consistently better power than existing approaches ., An application to a previously published dataset on Type-1 Diabetes 15 confirmed the original finding , namely that rare variants in IFIH1 confer protection towards disease ., The weights underlying our weighted-sum statistic depend only on the data at hand ., However , external information on the likelihood of a variant to be functional could prove very useful , and could be combined with the information present in the data to improve power to identify disease susceptibility variants ., Such information has been successfully used to identify the genes for several monogenic disorders 26 ., Price et al . 25 discuss a weighted-sum approach with externally-derived weights , and show that such information can be very useful using several empirical datasets ., We have also described a natural way to take into account such external functional predictions within the proposed framework ., Since empirical data are only now becoming available , it is not known how often both risk and protective variants are present in a particular disease gene ., When both types of variants are present , it seems appealing to be able to combine the two types of signals ., It is possible to extend the proposed approach to take advantage of both kinds of disease variants , and we discuss such an extension in Text S4 ., We noticed in our simulation experiments that such a hybrid approach can have much improved power when both types of variants are present , but this comes at the price of reduced power when only one type of variants is present ., Therefore , depending on the underlying disease model , both approaches could provide useful information ., The proposed approach is applicable to a case-control design and therefore is susceptible to spurious findings due to population stratification ., Population stratification has been shown to be an important issue in the context of common variants ., For rare variants , differences in rare variant frequencies between populations are likely to be even more pronounced ., Development of new methods , and extension of existing methods are necessary to adequately address the issue ., Alternatively , family-based designs offer the advantage of being robust to false positive findings due to population stratification ., Replication of association signals in independent datasets is an essential aspect of any disease association study , and has become standard practice for common variants ., Rare variants , due to their low frequencies and potential modest effects , are normally tested together with other rare variants in the same unit , e . g . , gene ., Therefore a reasonable first replication strategy is at the level of the gene ., Follow-up of individual variants in the gene can be performed to investigate whether any of the rare variants in the gene can be found to be significantly associated with disease ., Finding rare disease susceptibility variants is a challenging problem , especially due to their low frequencies and the probable moderate effects on disease ., So far the methods proposed in the literature have focused on case-control designs ., However , for rare variants , family-based designs may prove very useful ., Not only are they robust against population stratification , but they may also offer increased power due to the increased likelihood of affected relatives to share the same rare disease variants ., Continued development of novel statistical methods for identifying rare disease susceptibility variants is needed for population-based designs , and especially for family-based designs ., Software implementing these methods is available at: http://www . mailman . columbia . edu/our-faculty/profile ?, uni=ii2135 .

Introduction, Methods, Results, Discussion

Rapid advances in sequencing technologies set the stage for the large-scale medical sequencing efforts to be performed in the near future , with the goal of assessing the importance of rare variants in complex diseases ., The discovery of new disease susceptibility genes requires powerful statistical methods for rare variant analysis ., The low frequency and the expected large number of such variants pose great difficulties for the analysis of these data ., We propose here a robust and powerful testing strategy to study the role rare variants may play in affecting susceptibility to complex traits ., The strategy is based on assessing whether rare variants in a genetic region collectively occur at significantly higher frequencies in cases compared with controls ( or vice versa ) ., A main feature of the proposed methodology is that , although it is an overall test assessing a possibly large number of rare variants simultaneously , the disease variants can be both protective and risk variants , with moderate decreases in statistical power when both types of variants are present ., Using simulations , we show that this approach can be powerful under complex and general disease models , as well as in larger genetic regions where the proportion of disease susceptibility variants may be small ., Comparisons with previously published tests on simulated data show that the proposed approach can have better power than the existing methods ., An application to a recently published study on Type-1 Diabetes finds rare variants in gene IFIH1 to be protective against Type-1 Diabetes .

Risk to common diseases , such as diabetes , heart disease , etc . , is influenced by a complex interaction among genetic and environmental factors ., Most of the disease-association studies conducted so far have focused on common variants , widely available on genotyping platforms ., However , recent advances in sequencing technologies pave the way for large-scale medical sequencing studies with the goal of elucidating the role rare variants may play in affecting susceptibility to complex traits ., The large number of rare variants and their low frequencies pose great challenges for the analysis of these data ., We present here a novel testing strategy , based on a weighted-sum statistic , that is less sensitive than existing methods to the presence of both risk and protective variants in the genetic region under investigation ., We show applications to simulated data and to a real dataset on Type-1 Diabetes .

genetics and genomics/genetics of disease, genetics and genomics/complex traits, mathematics/statistics

journal.pgen.1003741

Genome Analysis of a Transmissible Lineage of Pseudomonas aeruginosa Reveals Pathoadaptive Mutations and Distinct Evolutionary Paths of Hypermutators

A molecular and mechanistic understanding of how bacterial pathogens evolve during infection of their human hosts is important for our ability to fight infections ., The advent of high-throughput sequencing techniques now offer unprecedented nucleotide resolution to determine the relatedness among infecting bacterial isolates and to unveil genetic adaptation within infected individuals and in response to antibiotic therapy 1–7 ., Unraveling the genetic content of pathogens helps to identify the genes that make certain bacterial lineages more pathogenic than others ., Nonetheless , the pathogenicity of a bacterial clone can also evolve via the mutational changes of pre-existing genes , a mechanism which is also known as pathogenicity- or pathoadaptive mutations 8 ., While several studies have provided insight into the genomic evolution of primary bacterial pathogens such as Yersinia pestis 6 and Vibrio cholera 2 causing acute infections , only little is known how these observations relate to opportunistic pathogens causing long-term infections 1 ., The opportunistic pathogen Pseudomonas aeruginosa is a common environmental inhabitant , which is also capable of causing both acute and chronic infections in a range of hosts from amoeba and plants to humans ., For example , P . aeruginosa causes chronic airway infections in most patients with cystic fibrosis ( CF ) , and is directly associated with the morbidity and mortality connected with this disease ., Chronic CF infections provide an opportunity for long-term monitoring of the battle between the infecting bacteria and the host immune defense and clinical intervention therapy 9 , 10 , and thus offer a direct method for observing evolutionary mechanisms in vivo ., In an effort to understand the evolutionary mechanisms facilitating the transition of P . aeruginosa from its environment to a human host , we have previously found no evidence for horizontal acquisition of genes to play a role 9 ., Instead , we suggested the establishment of long-term chronic infections to be a matter of tuning the existing genome via pathoadaptive mutations ., The within-host mutation rate is a key factor in determining the potential for bacterial pathogens to genetically adapt to the host immune system and drug therapies , and knowledge about in vivo growth dynamics of bacterial pathogens and their capacity for accumulation of mutations is essential for the design of optimal interventions ., Interestingly , the generation of mutations is frequently accelerated in clinical populations of P . aeruginosa that evolves as so-called ‘hypermutators’ due to deficient DNA mismatch repair systems 11 ., Although the hypermutable phenotype is also observed for other species in a range of conditions 12–16 , the impact of this phenotype in a natural environment and in relation to infections remains less clear ., Here we analyze the genome sequences of 55 isolates of the transmissible P . aeruginosa DK2 clone type causing chronic infections in a cohort of Danish CF patients ., Our collection , comprising both normal ( normomutator ) and hypermutable isolates , enabled a comparative analysis of evolutionary trajectories of individual sub-lineages of the DK2 clone type making it possible to identify genes targeted by pathoadaptive mutations ., Furthermore , the long-term population dynamics and structure of the clonal expanding DK2 lineage was elucidated by a high-resolution phylogeny , and an examination of the mutation dynamics of homopolymers ( homopolymeric tracts of identical nucleotides , e . g . GGGGG ) provided novel genome-wide evidence for the potential advantage of differential mutagenesis associated with the hypermutator phenotype ., We sequenced the genomes of a collection of 55 P . aeruginosa DK2 clones sampled from Danish CF patients between 1972 and 2008 ( Figure 1 and Table S4 ) ., The sequence data of 45 of the isolates have previously been reported 9 , 10 ., Most patients ( n\u200a=\u200a19 ) were represented by only a single or a few ( ≤4 ) isolates ., However , two patients were represented by 11 and 15 isolates ( CF173 and CF333 , respectively ) ., We identified a total of 7 , 326 unique SNPs in the 55 DK2 genomes , that could be explained by 7 , 368 mutational events ( consistency index 0 . 99 ) using a maximum-parsimonious phylogenetic model to elucidate the evolutionary relationship of the P . aeruginosa DK2 population ( Figure 2 ) ., The high consistency of the tree reflects the unidirectional , clonal evolution from the root of the tree to the tips , thus enabling inferences about the succession of mutations and the relationship among P . aeruginosa DK2 clones ., From the phylogenetic tree we observed a linear correlation between the number of SNPs and the time of sampling ( i . e . a constant rate of mutation accumulation during the clonal expansion of the DK2 lineage ) ( Figure 3 ) ., However , nine sub-lineages ( indicated by filled circles in Figure 3 ) deviated from this trend and had accumulated mutations at higher rates ., In one of these isolates , CF224-2002a , we found that 265 of the 273 SNPs accumulated in the branch leading to the isolate were densely clustered in two chromosomal regions with SNP densities ( 1 . 2 and 1 . 8 SNPs per kb , respectively ) , that are much higher than expected ( 0 . 043 SNPs per kb assuming a random distribution of SNPs ) ( Figure S1 ) ., The most likely explanation for these high SNP densities is that the two genomic regions are the result of recombination events with DNA from a P . aeruginosa strain ( s ) unrelated to the DK2 clone type ., Another study by Chung et al . observed similar indications of within-patient recombination events in P . aeruginosa 17 ., We found no evidence for additional recombination events among the 55 genome sequences ., The excess numbers of mutations in the remaining eight deviant isolates were the result of increased mutation rates due to mutations in mismatch repair and error prevention genes ., Seven isolates had non-synonymous mutations in one of the DNA mismatch repair ( MMR ) genes mutS ( n\u200a=\u200a2 ) and mutL ( n\u200a=\u200a4 ) or both ( n\u200a=\u200a1 ) , and their excess numbers of SNPs showed a highly increased transition∶transversion ratio consistent with MutS or MutL defects ( Table 1 ) 3 , 11 , 17–19 ., Moreover , one isolate ( CF173-1991 ) had a mutation in mutY and a molecular signature consistent with a MutY defect ( i . e . a high proportion of transversions ) ( Table 1 ) ., We did not find other mutations in mutS or mutL among the remaining genome sequences , but three additional early isolates ( CF84-1972 , CF43-1973 , and CF105-1973 ) had mutations in mutY as well as having the molecular signature associated with a MutY defect ( Table 1 ) ., In total , we found 11 hypermutator strains among the 55 isolates ., These mutators were found in ten of the 21 patients in our study ( 48% ) , which is comparable to previous findings ( 36% ) 11 ., Our results include two patients ( CF211 and CF224 ) from whom we isolated both hypermutators and normal ( normomutator ) clones documenting the co-existence of both types ., Indeed , the identification of both a hypermutable and a normal sub-lineage in years 1997 and 2006 from patient CF211 suggests at least 9 years of co-existence within this patient ( Figure 2 ) ., It is possible that the sub-lineages with different mutation rates occupy different niches within the hosts , each niche representing different selection pressures ., We next designed our mutational analysis to detect small insertions and deletions ( microindels ) ., A total of 1 , 204 unique microindels were discovered ., The inheritance was explained by 1 , 380 parsimonious events and was congruent with the SNP-based phylogeny although the consistency for the microindels was lower ( 0 . 87 ) than for the SNPs ( 0 . 99 ) ., The higher rate of homoplasy among microindels would a priori be expected as microindels accumulate with high rates at mutational hotspots consisting of simple sequence repeats ( SSRs ) 20 ., Accordingly , 93% of the inconsistent microindels were located in SSRs ., As expected from current knowledge 21 , 22 , we observed that the seven mutL/mutS hypermutators were particularly prone to mutation within SSRs consisting of homopolymers , and as a result 86% of the microindels that accumulated in the mutS/mutL hypermutable sub-lineages were localized in homopolymers whereas this was only true for 21% of the microindels within the remaining sub-lineages ( Table 1 ) ., Highly mutable loci have been shown to be important for pathogenesis and host adaptation of several pathogens 23 ., For example , increased mutation rates of homopolymers in MMR-deficient P . aeruginosa strains have been shown in vitro to be important for mutational inactivation of the regulatory gene mucA 24 , which is pivotal for adaptation in CF airways ., Nonetheless , we only have a limited understanding of the homopolymer mutation dynamics at a genome-wide level and of the impact of increased mutation rates of homopolymers in relation to host adaptation ., However , our collection of genome sequences from both normal and hypermutator isolates , sampled from the airways of CF patients , provides an opportunity to shed new light on homopolymer mutation rates and their impact on adaptation ., For each of the seven mutS/mutL hypermutator sub-lineages we calculated the mutation rates of homopolymers of different lengths ( Figure 4 ) ., We observed that longer homopolymers were more likely mutated than homopolymers of shorter lengths , and for homopolymers of 3–6 nucleotides length the mutation rate increased exponentially ( R\u200a=\u200a0 . 995; Students t-test , P\u200a=\u200a0 . 0026 ) ., One might expect large homopolymers to exhibit higher probabilities of mutation , because they are distributed more frequently outside coding regions ., However , we observed no evidence of this playing a role , as mutation rates of intergenic and intragenic homopolymers were similar ( Figure S2 ) ., Instead , the size-dependent mutation rate of homopolymers is likely to be a consequence of the mechanistically determined probabilities of strand-slippage during replication 20 ., The size-dependent mutation rates of homopolymers of different lengths suggest that different genes have different probabilities of mutation ., In this way , certain genes , in which variation is appreciated , may harbor sequences that are more frequently mutated in contrast to essential genes in which genetic changes are strongly selected against 23 ., In agreement with this , we find that genes annotated as essential genes in P . aeruginosa PAO1 25 are less likely to contain large homopolymers ≥7 nt ( Fishers exact test , P\u200a=\u200a0 . 037 ) ( Table S1 ) ., Survival of bacteria in human hosts has previously been suggested to be positively influenced by rapid modulation of the cell envelope ., In agreement with this ( and in opposition to essential genes ) , we find that genes functionally related with the composition of the cell envelope are more likely to contain large ≥7 nt homopolymers ( Fishers exact test , P\u200a=\u200a0 . 002 ) ( Table S1 ) ., This leads us to speculate that hypermutators have a selective advantage over their normal counterparts , not only because they can speed up evolution , but also because they are creating a bias towards a different evolutionary path by homopolymer facilitated differential mutagenesis ., In support of our hypothesis , we find that mutS/mutL hypermutators acquire 3 . 7 more mutations in cell envelope genes containing large homopolymers ( ≥7 nt ) relative to cell envelope genes without large homopolymers , and that the accumulation of mutations in the homopolymer-containing cell envelope genes is due to mutations within the homopolymers ., Accordingly , 50% of mutations in homopolymer-containing cell envelope genes are indels whereas this is only true for 5% of the mutations in the remaining genes ( 9/164 vs . 8/8; Fishers exact test , P\u200a=\u200a6 . 2×10−6 ) ., In further support of our hypothesis on differential mutagenesis we find two genes ( PADK2_15360 and PADK2_03970 ) in which all seven mutS/mutL hypermutators , but no other isolates , carries mutations ., Given the number of mutations within each of the seven hypermutable sub-lineages and all other lineages this observation is highly unexpected by chance ( P ( X≥2 ) ∼binom ( X; 5976; 2 . 9×10−7 ) =\u200a1 . 6×10−6; where 2 . 9×10−7 is the probability of an average length gene to be mutated in only the mutS/mutL sub-lineages ) ., One of the genes , PADK2_15360 , encodes an outer membrane receptor protein , and all seven hypermutators are independently mutated in the same 7×G homopolymer located at position 1127–1133 within the 2958 nt gene ., Since none of the other 48 isolates contain mutations within PADK2_15360 , we suggest that mutations in this gene represent a hypermutator-specific adaptive target for rapid modulation of the cell envelope ., All seven mutations are frameshift mutations causing premature stop codons resulting in truncated proteins without a putative TonB dependent receptor domain ( Pfam family PF00593 ) located in the C-terminal part of the protein ., We hypothesize that this domain is localized in the outer membrane where it , due to its potential surface-exposure , could be a target of recognition by the immune defense ., To further investigate the within-host evolutionary history of the DK2 lineage and to estimate the dates of divergence between DK2 isolates , we applied Bayesian statistics to infer time-measured phylogenies using a relaxed molecular clock rate model ( Figure 5 ) ., We excluded the hypermutator isolates and isolate CF224-2002a containing recombined regions from the analysis , as they would otherwise interfere with the phylogenetic analysis ., Based on this analysis , the mean mutation rate was estimated to be 2 . 6 SNPs/year ( 95% highest posterior density ( HPD; see Materials and Methods ) 1 . 8–3 . 2 SNPs/year ) which is equivalent to 4×10−7 SNPs/year per site or 9×10−11–11×10−11 SNPs/bp per generation assuming 3700–4500 generations per year 26 ., Our estimated mutation rate is in the same range as those estimated for Shigella sonnei ( 6×10−7 SNPs/bp/year ) 7 and Vibrio cholerae ( 8×10−7 ) 2 but in between the rates reported for Yersinia pestis ( 2×10−8 ) 6 and Staphylococcus aureus ( 3×10−6 ) 27 ., The topologies of the Bayesian phylogenetic reconstruction and the maximum-parsimonious phylogeny were congruent , and the relationship among the clones correlated with patient origin and the time of sampling ( Figure 2; Figure 5 ) ., We have previously shown that a set of specific mutations first observed in CF30-1979 and in all isolates sampled after 1979 were important for the reproductive success of the DK2 lineage and its dissemination among multiple individuals 10 ., Using the phylogenetic reconstruction , we estimate that isolates sampled after 1979 diverged from a common ancestor in year 1970 ( 95% HPD , 1961–1976 ) 10 ., Furthermore , our phylogenetic data document that the transmission potential of the DK2 lineage has been maintained over several decades ., The most recent transmission event is predicted to have occurred in year 1997 ( 95% HPD , 1991–2001 ) , as this is the latest time estimate of a predicted ancestor shared by isolates from different patients ( CF177-2002 and CF223-2002 ) ., Since we have not investigated DK2 isolates from all patients chronically infected with this lineage it remains a possibility that transmission has occurred subsequent to this time ., Seven patients are represented by multiple isolates , and in six of the patients at least two of the isolates clustered as monophyletic groups according to patient origin ( Figure 2; Figure 5 ) ., This is in agreement with a model in which independent sub-lineages of the DK2 clone evolved separately within individual patients , and it excludes the possibility of continuous and near-perfect mixing of strains between patients ., The patient-linkage was most prevalent for patient CF333 from which all 15 isolates constituted a single monophyletic group , and the isolates branched in general according to their sampling year giving a linear evolutionary trajectory with an average distance of 6 . 1 SNPs ( ∼2 . 3 years ) from the line of descent ( Figure 2; Figure 5 ) ., In contrast , we observed an unexpected DK2 population dynamics in patient CF173 in which the isolates clustered as three different monophyletic groups with four , five and two isolates , respectively ( Figure 2 ) ., This shows that patient CF173 was infected by three distinct sub-lineages rather than only a single sub-lineage ., Interestingly , the three sub-lineages carried by patient CF173 can be distinguished based on the sampling year of the isolates ., Accordingly , the isolates from the different clusters are sampled in the time-periods 1984–1991 ( cluster A ) , 1992–1999 ( cluster B ) and 2002–2005 ( cluster C ) , respectively ., This points to a replacement of the earlier sub-lineages around years 1991–1992 and 1999–2002 , respectively , caused by secondary transmission events ., Alternatively , it could be the result of co-existing lineages whose time-dependent sampling was caused by shifts in relative abundance or changes in sampling probability from different niches ., The presence of independently evolving DK2 sub-lineages made it possible to search for recurrent patterns of mutation and to identify bacterial genes that have acquired mutations in parallel in different individuals 1 , 28 ., Overall , we found no evidence for either intragenic bias of the mutations or for positive selection within coding regions ( dN/dS\u200a=\u200a0 . 66 including all mutations; Text S1 ) , and we would therefore expect the 7 , 383 intragenic mutations to be distributed randomly among the 5 , 976 P . aeruginosa DK2 genes ., This means that on average a gene would acquire 1 . 2 mutations , and we would expect only 1 . 3 genes to acquire mutations more than 6 times ( P ( X>6 ) ∼binom ( X; 7 , 383; 5976−1 ) =\u200a2 . 2×10−4 ) ., Nonetheless , we identified 65 genes that were mutated more than 6 times when comparing across all DK2 sub-lineages ( see Table S2 for the full list of all 65 genes ) ., The high mutation number within these genes could be the result of a positive selection for mutations , which is supported by our observation that increased pressures of selection acts on the top most mutated genes ( Figure 6 ) ., Accordingly , the signature for selection for SNPs accumulated in the 65 top most mutated genes ( dN/dS\u200a=\u200a1 . 11 ) was positive and significantly higher than for SNPs accumulated in other genes ( dN/dS\u200a=\u200a0 . 69; Fishers exact test , P\u200a=\u200a5 . 2×10−5 ) ., These findings suggest that the 65 genes with multiple mutations undergo adaptive evolution ( i . e . they are pathoadaptive genes involved in host adaptation ) , although the presence of neutral mutational hotspots or fast acquisition of secondary mutations within the same gene may contribute to the high mutation number in some genes ., To exclude the possibility that the high mutation numbers were the result of recombination events or because of particularly large gene sizes , we left out mutations from recombined regions and large genes ( >5 kb ) from our analysis ., A large part of the identified pathoadaptive genes were associated with antibiotic resistance ( n\u200a=\u200a14 ) , including the genes ampC , emrB , ftsI , fusA , gyrA/B , mexB/Y , pmrB , pprA , oprD , and rpoB/C ( Figure 7 and Table S2 ) , in which mutations have been shown to confer resistance against a range of antibiotics , e . g . beta-lactams , tetracyclines , quinolones , chloramphenicol , macrolides , fusidic acid , aminoglycosides , polymycins and penicillins 29–37 ., As such , the detection of multiple mutations in known antibiotic resistance genes confirmed the ability of our approach to identify genes involved in host adaptation ., The exact amino acid changes caused by nine out of 16 unique non-synonymous mutations found within the genes gyrA/B and rpoB have previously been shown to confer resistance against fluoroquinolones and rifampicin , respectively ( Table S3 ) ., Another major group of pathoadaptive genes ( n\u200a=\u200a18 ) were functionally related to the cell envelope ( Figure 7 and Table S2 ) ., Possibly , these mutations have been selected to evade the host immune response 38 or , especially in the case of lpxO2 , to prevent interaction from LPS-targeting antibiotics 39 ., Also , mutations in 13 genes involved in gene regulation were identified in our analysis , suggesting that remodeling of regulatory networks is a key evolutionary pathway in host adaptation as it seems to be in evolving Escherichia coli populations 40 ., Among the regulatory genes that acquired mutations were four yet uncharacterized genes encoding components of two-component regulatory systems , a gene-category which is significantly overrepresented ( 88/5823 vs . 7/58 Fishers exact test , P\u200a=\u200a6 . 7×10−5 ) among the pathoadaptive genes ( Table S2 ) ., We suggest that these uncharacterized regulatory genes as well as other genes identified as involved host adaptation represent potential therapeutic targets ., The adaptive benefits of a mutation are usually investigated by introduction of single or multiple mutations into isogenic strains and testing for fitness effects associated with the mutation ( s ) in controlled experimental conditions ( such as competition experiments ) ., Such testing is most effective when the phenotype ( e . g . antibiotic resistance ) can be easily interpreted in relation to the fitness impact ., However , for mutations for which no or only subtle phenotypic changes are apparent it is difficult to directly test the fitness effects ., In addition , the impacts on fitness of specific mutations must be assessed in the same environment as the one in which the mutation was selected ., This is obviously not possible in case of human airway infections ., To circumvent these limitations , we hypothesize that the count of mutations within the pathoadaptive genes can be used as a measure of the fitness of individual clones of P . aeruginosa ., To investigate this hypothesis we took advantage of the two strain displacements ( or changes in strain abundances ) that occurred in patient CF173 in the years 1991–1992 and 1999–2002 , which suggested that CF173 was infected by three succeeding DK2 sub-lineages A ( 1980–1991 ) , B ( 1990–1999 ) , and C ( 2000–2005 ) ., We assume that the succeeding sub-lineage must be better adapted ( i . e . having a higher fitness ) than the previous sub-lineage , which was outcompeted ., When determining the number of mutations found in the sub-lineages A , B , and C within the pathoadaptive genes , it was striking that the succeeding genotype consistently had a higher count of mutations than the previous genotype ( Figure 7 ) ., In this way , the counts of mutations correlated with the strain displacement observed within patient CF173 ., We suggest that the mutation count can be used to predict the fitness of emerging DK2 clones , and that the pathogenicity scoring together with the information about the specific mutations can be used as a novel approach for clinicians to treat and segregate patients ., It should be noted that our results cannot simply be ascribed to the succeeding genotypes having more mutations in general as no significant positive correlation existed between the total number of mutations and the number of mutations within the pathoadaptive genes ( R\u200a=\u200a0 . 30; Students t-test , P\u200a=\u200a0 . 28 ) ., By genome sequencing of 55 isolates of the transmissible DK2 clone type of P . aeruginosa , we have provided a detailed view of the evolution of a bacterial pathogen within its human host ., The sampling from multiple patients offered the opportunity to detect loci that were independently mutated in parallel lineages , here referred to as pathoadaptive genes , whereas sampling multiple times from the same patient gave an opportunity to study the within-patient population dynamics ., Several of the pathoadaptive genes identified here were associated with antibiotic resistance , gene regulation , and composition of the cell envelope ., Some of these genes have been found in other studies of genomic evolution in CF pathogens to be important for adaptation 1 , 3 , 17 ., Genomic analysis of additional P . aeruginosa lineages from different patients and clinical settings will enable a systematic identification of genes that are repeated targets for selective mutations during adaptation to life in the CF lung ., Importantly , we also identified genes of unknown function and without prior implication in pathogenesis ., Further investigations of the function of these genes are required to determine their potential as future therapeutic targets against the infection ., An exceptional 21-year time series of 11 isolates sampled from patient CF173 revealed a complex population dynamics in which the patient was infected by three distinct sub-lineages of the DK2 clone type , each sub-lineage being dominant over several years until its final decline or disappearance ., This observation illustrates the power of high-throughput sequencing in relation to uncovering pathogen dynamics within infected individuals ., We further observed that the cumulative count of mutations within pathoadaptive genes increased for each of the succeeding sub-lineages ., This means that emerging sub-lineages carried a cumulative palette of pathoadaptive mutations and not only adaptive mutations conferring an advantage for a newly introduced selection force that may have triggered the removal of the preceding lineage ., The identification of pathoadaptive genes involved in host adaptation and our finding that the specific count of mutations within these genes act as a classifier that predict the pathogenicity of emerging sub-lineages of the DK2 clone type , should enable better epidemiological predictions and provide valuable information for the clinicians on how to treat and segregate patients ., The presence of hypermutable lineages within 48% of the studied individuals might be the outcome of an accelerated acquisition of beneficial mutations within hypermutators 11 , 41–43 ., Nonetheless , our examination of mutation dynamics of homopolymers provided a novel genome-wide perspective on the impact and potential advantage of differential mutagenesis associated with the hypermutator phenotype ., Showing a clear exponential correlation between the rate of change and the size of the homopolymer , we confirmed homopolymers to be hotspots for differential mutagenesis , and we identified two homopolymer-containing genes to be preferentially mutated in hypermutators ., In conclusion , we have shown how collections of isolates of bacteria sampled from chronically infected patients constitute a valuable basis for studying evolution of pathogens in vivo , and our results facilitates comparative studies as sequencing datasets become increasingly available ., The study encompasses 55 isolates of the P . aeruginosa DK2 clone type that were sampled over 38 years from 21 CF patients attending the Copenhagen Cystic Fibrosis Center at the University Hospital , Rigshospitalet ( Figure 1 ) ., Isolation and identification of P . aeruginosa from sputum was done as previously described 44 ., Sequencing of 45 of the isolates was previously reported by Yang et . al . 10 and Rau et al . 9 ., Two of the previously sequenced isolates ( CF333-1991 and CF510-2006 ) were re-sequenced together with ten new isolates on an Illumina HiSeq2000 platform generating 100-bp paired-end reads using a multiplexed protocol to an average coverage depth of 63–212 fold ., Sequence reads from all isolates are deposited in the Short Read Archive under accession number ERP002277 ( accession numbers for individual samples are provided in Table S4 ) ., Reads were mapped against the P . aeruginosa DK2 reference genome ( CF333-2007a; Genbank accession no . CP003149 ) using Novoalign ( Novocraft Technologies ) 45 , and pileups of the read alignments were produced by SAMtools release 0 . 1 . 7 46 ., Single nucleotide polymorphisms were called by the varFilter algorithm in SAMtools in which minimum SNP coverage was set to 3 ( samtools . pl varFilter -d 3 -D 10000 ) ., Only SNP calls with quality scores ( Phred-scaled probability of sample reads being homozygous reference ) of at least 50 ( i . e . P≤10−5 ) were retained ., Microindels were extracted from the read pileup by the following criteria; ( 1 ) quality scores of at least 500 , ( 2 ) root-mean-square ( RMS ) mapping qualities of at least 25 , and ( 3 ) support from at least one fifth of the covering reads ., The false-negative rates were found to be 2% and 3% by in silico introduction of random base-substitutions and microindels ( lengths 1–10 bp ) , respectively ., To avoid false-positives , the reference genome was re-sequenced by Illumina sequencing to exclude polymorphisms caused by errors in reference assembly ., Also , Illumina re-sequencing of CF333-1991 confirmed all the SNPs ( and found no other SNPs ) that were previously reported for this isolate by use of pyrosequencing 10 ., Indeed , the confirmation by re-sequencing of CF333-1991 and the fact that many isolates are only discriminated by a few mutations verify that our genomic analysis has a very low false-positive rate ., A maximum-parsimonious phylogenetic analysis was used to predict the relationship and mutational events among the clones of the DK2 clone type ., The tree consistency index ( CI\u200a=\u200am/s ) was calculated as the minimum number of changes ( m ) divided by the number of changes required on the tree ( s ) ., The CI will equal 1 when there is no homoplasy ., For the calculation of average distances of the 15 CF333 isolates to their line of descent , the line of descent was defined as the direct lineage from the most recent common ancestor ( MRCA ) of all 15 isolates until the MRCA of the three most recently sampled isolates ( CF333-2007a , CF333-2007b , CF333-2007c ) ., To provide the most accurate estimates of the relative homopolymer mutation rates in the mutS/mutL MMR-deficient sub-lineages , we calculated the rates per mutS/mutL MMR-deficiency caused SNP ., This corrected count of SNPs were found by subtracting the fraction of SNPs expected to have accumulated due to the normal underlying mutation rate , i . e . SNPs not caused by the mutS/mutL MMR-deficiency ., For this purpose a 2∶1 transition to transversion ratio was assumed for the normal background mutation rate ., This means that the SNP count of hypermutator branch “KD” composed of 2 , 534 SNPs ( Table 1 ) , hereof 29 transversions , was corrected to 2 , 447 mutS/mutL MMR-deficiency caused SNPs ., All results and conclusions were unaffected from this correction ., Bayesian analysis of evolutionary rates and divergence times was performed using BEAST v1 . 7 . 2 47 ., BEAST was run with isolate CF510-2006 as an outgroup 9 and the following user-determined settings; a lognormal relaxed molecular clock model which allows rates of evolution to vary amongst the branches of the tree , and a general time-reversible substitution model with gamma correction ., Results were produced from three independent chains of 50 million steps each , sampled every 5 , 000 steps ., The first 5 million steps of each chain were discarded as a burn-in ., The results were combined , and the maximum clade credibility tree was generated ( using LogCombiner and TreeAnnotator programs from the BEAST package , respectively ) ., The effective sample-sizes ( ESS ) of all parameters were >500 as calculated by Tracer v1 . 5 ( available from http://beast . bio . ed . ac . uk/Tracer ) , which was also used to calculate 95% HPD confidence intervals of the mutation rate ( i . e . an interval in which the modeled parameter resides with 95% probability ) ., The root of the tree was predicted to be in year 1943 ( 95% HPD , 1910–1962 ) ., Note

Introduction, Results and Discussion, Materials and Methods

Genome sequencing of bacterial pathogens has advanced our understanding of their evolution , epidemiology , and response to antibiotic therapy ., However , we still have only a limited knowledge of the molecular changes in in vivo evolving bacterial populations in relation to long-term , chronic infections ., For example , it remains unclear what genes are mutated to facilitate the establishment of long-term existence in the human host environment , and in which way acquisition of a hypermutator phenotype with enhanced rates of spontaneous mutations influences the evolutionary trajectory of the pathogen ., Here we perform a retrospective study of the DK2 clone type of P . aeruginosa isolated from Danish patients suffering from cystic fibrosis ( CF ) , and analyze the genomes of 55 bacterial isolates collected from 21 infected individuals over 38 years ., Our phylogenetic analysis of 8 , 530 mutations in the DK2 genomes shows that the ancestral DK2 clone type spread among CF patients through several independent transmission events ., Subsequent to transmission , sub-lineages evolved independently for years in separate hosts , creating a unique possibility to study parallel evolution and identification of genes targeted by mutations to optimize pathogen fitness ( pathoadaptive mutations ) ., These genes were related to antibiotic resistance , the cell envelope , or regulatory functions , and we find that the prevalence of pathoadaptive mutations correlates with evolutionary success of co-evolving sub-lineages ., The long-term co-existence of both normal and hypermutator populations enabled comparative investigations of the mutation dynamics in homopolymeric sequences in which hypermutators are particularly prone to mutations ., We find a positive exponential correlation between the length of the homopolymer and its likelihood to acquire mutations and identify two homopolymer-containing genes preferentially mutated in hypermutators ., This homopolymer facilitated differential mutagenesis provides a novel genome-wide perspective on the different evolutionary trajectories of hypermutators , which may help explain their emergence in CF infections .

Pseudomonas aeruginosa is the dominating pathogen of chronic airway infections in patients with cystic fibrosis ( CF ) ., Although bacterial long-term persistence in CF hosts involves mutation and selection of genetic variants with increased fitness in the CF lung environment , our understanding of the within-host evolutionary processes is limited ., Here , we performed a retrospective study of the P . aeruginosa DK2 clone type , which is a transmissible clone isolated from chronically infected Danish CF patients over a period of 38 years ., Whole-genome analysis of DK2 isolates enabled a fine-grained reconstruction of the recent evolutionary history of the DK2 lineage and an identification of bacterial genes targeted by mutations to optimize pathogen fitness ., The identification of such pathoadaptive genes gives new insight into how the pathogen evolves under the selective pressures of the host immune system and drug therapies ., Furthermore , isolates with increased rates of mutation ( hypermutator phenotype ) emerged in the DK lineage ., While this phenotype may accelerate evolution , we also show that hypermutators display differential mutagenesis of certain genes which enable them to follow alternative evolutionary pathways ., Overall , our study identifies genes important for bacterial persistence and provides insight into the different mutational mechanisms that govern the adaptive genetic changes .

journal.pgen.1006051

Selective Retention of an Inactive Allele of the DKK2 Tumor Suppressor Gene in Hepatocellular Carcinoma

Hepatocellular carcinoma ( HCC ) is the fifth most common cancer worldwide , and the third leading cause of cancer-related mortality , contributing to over 660 , 000 annual deaths worldwide 1 , 2 ., HCC exhibits a distinct geographic distribution of over 80% of HCC cases occurring in Southeast Asia and sub-Saharan Africa ., It should also be noted that the incidence of HCC has recently increased significantly in the United States of America 3 ., Late-stage HCC cases typically display more genetic alterations than hyperplasia or dysplasia lesions; these alterations include chromosomal instability , DNA rearrangements , DNA methylation , and DNA hypomethylation 4 ., Several studies have identified recurrent chromosomal instability regions associated with HCC by comparative genomic hybridization ( CGH ) or loss of heterozygosity ( LOH ) mapping 5–10 ., The chromosomal gain regions involve 1q , 5q , 6p , 8q , 10q , 11q , 17q , and 20q , while the chromosomal loss regions involve 1p , 4q , 6q , 8p , 10q , 13q , 16q , and 17p 11 ., Several cancer genes have been identified and validated in these chromosomal instability regions ., However , the mechanisms by which these genomic alterations at multiple chromosomal segments of potential oncogenes and tumor suppressor genes lead to hepatocarcinogenesis remain undetermined ., The Wnt/β-catenin pathway is involved in homeostasis , cell proliferation , differentiation , motility , and apoptosis 12 ., Activation of the Wnt/β-catenin pathway frequently occurs in HCC 13 , 14 ., β-catenin overexpression and mutations related to this have been described during early-stage HCC development and HCC progression 15–17 ., More β-catenin mutations are manifested in hepatitis C virus-associated HCC than in hepatitis B virus-related HCC 17–19 ., It is interesting that β-catenin mutations are typically seen in HCC with a low-level genomic instability 20 , indicating that the Wnt/β-catenin pathway could represent an alternative route to hepatocarcinogenesis ., Accumulation of β-catenin in the nucleus has been observed in 40% to 70% of HCC cases 10 , 21 ., Several secreted proteins are known to negatively regulate the Wnt/β-catenin pathway ., These Wnt antagonists can be divided into two functional classes 22 ., One involves the Wise , sclerostin and Dickkopf ( DKK ) families that bind directly to LRP5/6 ., The other consists of Wnt inhibitory factors and secreted frizzled-related proteins that bind directly to soluble Wnt ligands ., The DKK family consists of secreted proteins that contain two cysteine-rich domains 23 and of four members ( DKK1 to DKK4 ) that are able to inhibit the Wnt co-receptors LRP5/6 and Kreman 1/2 24 , 25 ., Down-regulation of the DKK family , when observed in HCC , usually involves epigenetic inactivation either by methylation or via silencing by miRNA 22 , 26 ., On the basis of CGH and LOH studies , approximately 30% to 70% of HCC patients showed genetic alterations in bands 21–25 of chromosome 4q 27–29 ., Chromosome 4q21-25 loss is involved in early HCC development 29 ., To delineate the LOH pattern in chromosome 4q22-25 , we used ten STR markers from 92 . 5 Mb to 117 . 5 Mb on human chromosome 4 to determine the minimal critical region of LOH for 47 HCC cases ., As shown in Fig 1 , 28 cases ( 59 . 6% ) were determined to have LOH within chromosome 4q22-25 region , while the other cases were either non-informative or heterozygous ., The result is consistent with the overall LOH frequencies for chromosome 4q22-25 obtained from other studies ., According to the Knudsen’s two-hit theory 30 , cancer develops when a tumor suppressor gene mutation occurs in one allele , followed by the loss of the other allele , reflecting as LOH in the genetic analysis ., Thus , detection of variant sequences specifically associated with LOH in the tumor tissue is one method of identifying candidate tumor suppressor genes ., We have taken a re-sequencing approach in an attempt to discover significant sequence variations in the genes on chromosome 4q21-25 ., A total of 2 , 293 pairs of primers were designed for PCR to amplify target sequences; these include 1 , 449 exonic and upstream promoter regions of 152 known and predicted genes that reside in the interval from 76 . 8 Mb to 114 Mb on human chromosome 4 ( NCBI , build 33 ) ., In the pilot study using a sample panel consisting of 12 HCC patients and 12 healthy human controls , we identified a total of 1 , 574 sequence variations , consisting of 1 , 462 substitutions , 43 insertions , and 69 deletions ., Among these variations , 99 sequence variations of 62 genes were found to be significantly associated with HCC ( p < 0 . 05 ) ( S1 Table ) ., Using allelic retention status in the HCC tumor as a criterion , three genes ( UNC5C , DKK2 , and ZGRF1 ) from the LOH region were evaluated for further investigation ( Fig 2 ) ., UNC5C , which encodes ntetrin-1receptor , has been reported to function as tumor suppressor gene in human colon cancer 31 , 32 ., The DKK family is able to inhibit the Wnt signaling pathway in several cell types and is usually down-regulated in several different cancers 33 ., ZGRF1 , whose identity and function were not yet known at the time that we conducted the genotype analysis , is now grouped as a zinc finger gene in the database ., Interestingly , 6 of the 12 cases showed LOH in the ZGFR1 sequence and the tumors invariably retained the G-A-C-G haplotype for the four SNPs ., Of these variations that were associated with HCC , four that belong to the human DKK2 gene were of particular interest due to their location within the regulatory region of the gene; these consisted of three in the promoter region ( g . -967A>T , g . -923C>A , and g . -441T>G ) and one in the 5’UTR ( c . 550T>C ) ., To further investigate the association with HCC , we increased the subject number to 47 HCC cases and 88 healthy controls to analyze these four variations ., The results are summarized in Table 1 ., The association remained significant for DKK2_-967 , DKK2_-923 , and DKK2_+550 ( p < 0 . 05 ) ., Note that the two SNPs at the promoter region ( nucleotides positions -967 and -923 ) are in linkage disequilibrium , therefore , the allele frequency is the same between the two sites ., Genetic studies based on haplotypes have provided greater statistical power than those based on the underlying SNPs 34 ., To investigate whether or not there were specific DKK2 haplotypes that are associated with HCC , we determined the haplotypes of 88 healthy controls using GENECOUNTING 2 . 2 ., A total of four haplotypes that had a probability higher than 0 . 02% were predicted ( Table 2 ) ., Among them , two major haplotypes–haplotype 2 ( ACTT ) and haplotype 3 ( TATT ) –had a combined frequency of nearly 76% in the studied subjects; haplotype 2 was the dominant haplotype ( 44 . 9% ) ., We also performed direct sequencing to determine the haplotypes of individually cloned genomic DNA fragments from the blood , tumor adjacent tissue , and tumor tissue of 16 HCC patients who were heterozygous for DKK2 ., Of 13 HCC cases , eight haplotypes were identified , including four recombinant haplotypes: haplotype 5 ( TAGT ) , haplotype 6 ( ACTC ) , haplotype 7 ( TATC ) and haplotype 8 ( ACGT ) ., Interestingly , these additional haplotypes were only detected in the non-neoplastic tissues but were absent from both the blood samples and tumor tissue samples ( Table 3 ) ., Notably , haplotype 1 ( TAGC ) was the most frequently observed haplotype in the tumor tissue samples from these HCC cases , observed in 13 out of 16 samples ., When we compared the haplotypes of blood , tumor adjacent tissue and tumor tissue from the same patients , we unexpectedly found that there were more than two haplotypes in the tumor adjacent tissues ., However , there was only one major allele , haplotype 1 , retained in the tumor tissue ., These results indicated that there had been frequent recombination events affecting DKK2 during HCC tumorigenesis and that the DKK2 haplotype 1 had been selectively retained in the tumors ., Three of the four identified DKK2 SNPs were located in the promoter region of this gene , while the fourth was in the 5’UTR ., We speculated that the various DKK2 haplotypes might show differences in transcriptional activity ., To address this issue , we measured the reporter activity of a luciferase gene that was driven by the promoter sequences of the DKK2 haplotype alleles ., A promoterless construct was used as a negative control , and the transcriptional activity was normalized against the transfection efficiency determined by β-galactosidase activity ., Haplotype 2 ( ACTT ) , which was found most frequently in the healthy controls , drove the expression of luciferase at a rate that was 10 fold higher than that of the promoterless construct ( Fig 3A ) ., Similarly , haplotype 3 ( TATT ) , which is referred to as wild type in the NCBI public database , drove the expression of luciferase at a rate 8 fold higher than the reference level ., In contrast , haplotype 1 ( TAGC ) was found most frequently in the tumor samples and showed significantly lower transcriptional activity when compared to the other seven haplotypes ( p < 0 . 001 ) ., To demonstrate that DKK2 haplotypes effect DKK2 expression and to confirm observations from in vitro studies , relative DKK2 expression levels between the tumor tissues and the non-tumor counterparts from 30 pairs HCC samples were analyzed by reverse transcription quantitative PCR ( RT-qPCR ) ., The relative expression levels were classified into three categories , determined by the presence of chromosome 4q24-25 LOH and/or DKK2 TAGC haplotype ( Fig 3B ) ., The difference between the two groups , non-LOH and LOH without TAGC , was not significant ( p = 0 . 229 ) ., However , the cohort with both chromosome 4q24-25 LOH and DKK2 TAGC haplotype showed significantly lower DKK2 expression levels than the other two cohorts: without chromosome 4q24-25 LOH ( p < 0 . 001 ) and with chromosome 4q24-25 LOH but no DKK2 TAGC haplotype ( p < 0 . 001 ) ., Taken together with the genotyping data , our results indicate that this transcriptionally inactive DKK2 allele was being selectively retained in the tumor when heterozygous HCC patients exhibited a LOH during tumorigenesis ., To investigate the function of DKK2 as part of the Wnt/β-catenin signaling pathway in hepatocytes , we incubated HuH-7 cells that had been transiently transfected with the TCF reporter plasmid with variable amounts of recombinant Wnt3a and DKK2 ., The plasmid contains multiple TCF binding sites upstream of the promoter , and the luciferase activity within the cells reflected the β-catenin concentration in the nucleus 35 ., As shown in Fig 4A , luciferase gene expression was correlated with Wnt3a concentration in a dose-dependent manner ( p < 0 . 05 ) ., With Wnt3a stimulation , there was significant association between the luciferase activity and DKK2 concentrations above 200 ng/ml ( p < 0 . 05 ) and DKK2 down-regulated Wnt3a-enhanced luciferase gene expression in a dose-dependent manner ( p < 0 . 05 ) ., This effect was paralleled between the luciferase assay and the cell proliferation assay ( Fig 4A and 4B ) ., Consistently , by abrogating the Wnt and receptor interaction at the cell surface , DKK2 inhibited β-catenin translocation from the cytosol to the nucleus ( Fig 4C ) ., The data confirmed that signaling molecules of the Wnt/β-catenin pathway are involved in oncogenesis by controling cell proliferation 36 ., Thus , the results of our DKK2 functional studies are consistent with previous reports whereby members of the DKK family are able to play a role in development and disease by modulating the Wnt/β-catenin pathway 22 ., To address possible mechanisms of LOH for the DKK2 gene , we analyzed the cytogenetic changes in eight HCC cases that were heterozygous for DKK2 haplotype 1 in their tumor adjacent tissue ., All eight HCC cases had chromosomal deletions of band 4q2l , and six cases showed LOH for 4q22-25 ( S2 Table ) ., Interestingly , three of the cases were polysomic and one was disomic for chromosome 4 with loss of 4q21 , as determined by dual-color FISH ., An example is shown in Fig 5 ., Sequencing of the DKK2 gene in the tumor tissue of these cases indicated that only haplotype 1 was retained in the tumor tissue , regardless of the copy number of 4q21 signals ., These results support the idea that , during HCC tumorigenesis , chromosome amplification occurs at the DKK2 locus prior to LOH ( Fig 6 ) ., In this study , we have taken a genetic approach to investigate the LOH region of human chromosome 4 and its role in HCC oncogenesis ., By scrutinizing the genetic variants in a 37 . 2 Mb region of common chromosomal loss that affects nearly 60% of the HCC cases , we have uncovered the tumor suppressor function of DKK2 in the liver ., Additionally , our study provides new insights regarding LOH in HCC ., First , we have shown that DKK2 function was compromised in HCC by the removal of active DKK2 alleles ., The Wnt signaling pathway plays an important role in liver cancer , and extensive studies have revealed that Wnt antagonists can be inactivated by epigenetic modification of the DKK coding genes 22 , 26 ., By way of contrast , our finding provides a new mechanism whereby DKK2 loses its function through selective retention of an inactive allele ( Fig 6 ) ., Thus , our data supports that this principle is also applicable to hepatocarcinogenesis ., Secondly , re-sequencing the LOH region allowed us to discover functional variants associated with hepatocarcinogenesis ., By detecting the differential distribution of haplotypes between blood , non-tumor tissue , and tumor tissue ( Table 3 ) , we were able to identify significant genetic changes in the chromosomal regions showing genomic instability ., Selective retention of a functional allele , in theory , could also give rise to overexpression of an oncogene ., Allelic imbalance in combination with DNA amplification has been detected in the HCC genome ., Given the frequent and extensive genomic changes associated with HCC , other tumor suppressor genes might also be inactivated through a similar mechanism ., For example , UNC5C is a known tumor suppressor gene 31 , 32 ., Within the 4q21-25 region , UNC5C displayed a nonrandom distribution of alleles in the HCC tumors when LOH has occurred ., The functional significance of the ZGFR1 gene showing LOH is currently unknown ., Thirdly , by taking a comprehensive approach on a focused region , our analysis revealed that there was hyper-recombination in the promoter region of the DKK2 sequence ( Table 3 ) ., We identified more than two haplotypes in the adjacent non-tumor liver tissues , yet most HCC cases retained haplotype 1 in the tumor tissues ., Myers et al . ( 2008 ) reported that two DNA motifs are associated with recombination hot spots: the 7-mer CCTCCCT and the 13-mer CCNCCNTNNCCNC; these are clustered in breakpoint regions and act as a driver of genome instability 37 ., We scanned the DKK2–1 . 5 kb to +1 kb region and found two CCTCCCT motifs in the DKK2 exon 1 sequence at +640 to +646 and +644 to +650 ( S1 Fig ) ., Furthermore , we searched ReDB ( http://www . bioinf . seu . edu . cn/ReDatabase/ ) , a recombination rate database to investigate the DKK2 locus ., Interestingly , the recombination rate of the DKK2–437 to -4 , 276 promoter region was dramatically increased from average of 0 . 02% to 14 . 73% ( S3 Table ) ., Thus , the results of the sequence analysis support the scheme shown in Fig 6 ., As the cell proliferation rate is elevated in the pre-cancerous tissues , DNA breakage is likely to occur near the recombination hotspots in the DKK2 promoter region and this will lead to loss of DKK2 alleles ., At the same time , those cells with low transcriptional activity of the DKK2 haplotype 1 allele are selected for clonal amplification during tumorigenesis ., Finally , our genetic and functional data confirms that DKK2 functions as a tumor suppressor in the liver ., The results from the functional analysis using cultured liver cancer cell support the hypothesis that DKK2 acts through the canonical Wnt pathway and antagonize the cell proliferation elicited by the Wnt3a ligand ( Fig 4 ) ., While this study was in progress , others studying different cancer types have reported that DKK2 functions as a tumor suppressor gene 38–40 ., Of particular relevance to liver cancer , Maass et al . ( 2015 ) recently published that a Dkk2 deletion in mice is associated with liver carcinogenesis and enrichment of stem cell properties 41 ., Thus , DKK2 might work through both Wnt-dependent and independent mechanisms during hepatocarcinogenesis ., Considering the role of DKK2 in HCC oncogenesis , genes affected by DKK2 modulation could possibly serve as biomarkers in epidemiological studies ., Additional work is warranted to address the implications of these findings with respect to disease classification and clinical management ., The study was approved by the Research Ethics Committee of National Health Research Institutes ( Permit Number: EC1030201-E ) and informed consent was obtained from each participant ., Human subjects were recruited from the Koo Foundation Sun Yat-Sen Cancer Center and Chang Gung Memorial Hospital ., These human specimens were collected under informed consent in accordance with the recommendations of Research Ethics Committee of National Health Research Institutes ., Genomic DNA and total RNA were isolated using the single-step method 42 from tumor tissues of the HCC patients as well as from their adjacent non-tumor tissues that appeared normal ., Primers specifically targeting each genomic fragment were designed using Primer3 ., Primer sequence information on the 2 , 293 amplicons is available on request ., PCR was initiated at 95°C for 10 minutes , followed by 45 cycles of 95°C for 30 seconds , annealing at various temperatures as appropriate to the primer pair for 30 seconds , and extension at 72°C for 45 seconds ., The final step was at 72°C for 3 minutes ., The optimal annealing temperature for each pair of primer was pre-tested ., The PCR products were treated with exonuclease I in order to remove unreacted primers ., DNA sequencing reactions were performed using Dye-terminator ( Applied Biosystems Inc . , Foster , CA ) and the same primers were used for the PCR amplification ., The products were separated by electrophoresis on an automated ABI 3700 PRISM DNA sequencer to determine the sequence of amplified fragments ., The results were analyzed using Phrap-Phred and PolyPhred ( ver . 10 ) software 43 ., Heterozygous variations were identified by the presence of double peaks at single nucleotide positions ., The forward primer 5-TTTGCTTGGAAAGTCTCGC-3 and the reverse primer 5-AGGGGTGGGAATGCAAAG-3 were used for PCR amplification of the -1 , 135 to +667 genomic region of the DKK2 gene ., The PCR products were subjected to TA cloning using the pGEM-T vector ( Promega ) ., After transformation , 96 colonies were individually selected for direct sequencing ., A DNA fragment , -1 , 135 to +667 of the DKK2 gene , was amplified using genomic DNA from each of the HCC cases with different haplotypes ., Sequence of the PCR product was verified before cloning into the pGL3 vector ., In total , 4 μg of pGL3-DKK2 promoter plasmid DNA and 0 . 8 μg of pcDNA3 . 1-His-LacZ plasmid DNA were co-transfected into HuH-7 cells ., After 48 hours , the cells were lyzed and the luciferase activity was detected by LucLite Kit ( Packard BioScience ) following the manufacture’s instruction ., To report the relative activity , the measured luciferase activity was normalized against the activity of β-galactosidase activity , which served as a transfection control ., Relative DKK2 expression levels between the tumor ( T ) tissues and the non-tumor ( N ) counterpart were determined using RT-qPCR ., Total RNA from 30 pairs of HCC samples were reverse-transcribed to cDNA using SuperScriptII ( Invitrogen ) according to the manufacturers instructions ., Subsequent qPCR reactions for DKK2 and β-actin were performed in triplicates on ABI StepOne real-time PCR system , using KAPA SYBR FAST ABI Prism 2X qPCR Master Mix ( Kapa Biosystems ) ., The sequences of the primers used for RT-qPCR were as follows: for DKK2 , 5’- GCAATAATGGCATCTGTATC ( forward ) and 5’- GTCTGATGATCGTAGGCAG ( reverse ) and for β-actin , 5’- ATCCGCAAAGACCTGTAC ( forward ) and 5’- GGAGGAGCAATGATCTTG ( reverse ) ., All samples were analyzed and normalized with expression level of the internal control gene , β-actin ., Relative quantification of fold-change was performed , comparing △CT of tumor tissues and △CT of tumor adjacent tissues ., For the TOPflash assay 35 , 2 μg of TCF reporter plasmid DNA and 0 . 5 μg of pcDNA3 . 1-His-LacZ plasmid DNA were co-transfected into HuH-7 cells ., After 24 hours , the cells were starved with DMEM medium containing 0 . 1% FBS for another 24 hours ., Then , the cells were cultured for 48 hours with medium that contained Wnt3a and/or DKK2 recombinant protein ( ng/ml ) ( Peprotech ) ., The TOPflash activity was measured by luciferase activity using the Dual-Luciferase Reporter Assay Kit ( Promega ) ., The data was normalized against β-galactosidase activity ., HuH-7 cells were plated in the 24 well plates ( 2x104 cells per well ) for 24 hours before the cells underwent serum starvation ., After 24 hours , the cells were cultured with DMEM medium containing Wnt3a and/or DKK2 recombinant protein ( ng/ml ) for 48 hours ., The cell proliferation assay was performed using alamarBlue cell viability reagent ( Thermo Scientific ) according to the user manual ., HuH-7 cells were serum-starved and stimulated with Wnt3a and/or DKK2 recombinant protein , as described above ., But , after 6 hours , the nucleus and cytoplasm were separated using the ProteoJET Cytoplasmic and Nuclear Protein Extraction Kit ( Fermentas ) ., Protein samples , loaded with 20 μg per lane , were separated by electrophoresis on a 10% SDS-PAGE gel and transferred onto a membrane ., Then , the membrane was probed with primary antibodies at optimal dilutions , followed by secondary antibody detection ., The primary antibodies used for the current study were anti-β-catenin ( Cell Signaling ) and anti-GAPDH ( Novus Biologicals ) and anti-Histone H3 ( Cell Signaling ) ., Touch slide preparations , probe preparations and fluorescence in situ hybridization were performed according to published protocols 44 ., In brief , a biotin-labeled 964 a_2 YAC probe specific to chromosome band 4q2l was cohybridized with a digoxigenin-labeled centromeric probe for chromosome 4 ., Signal detection was accomplished using avidin-FITC and rhodamine antidigoxigenin ., Nuclear counterstaining was carried out using 0 . 1 μg/ml DAPI in antifade solution ., To confirm significance of the data obtained from in vivo studies , Kruskal-Wallis H test was implemented to determine if the clusters were significantly different ., After significance was established , Mann-Whitney U tests were used to identify which cluster exhibited the greatest significance ., For in vitro data , variance pre-test was analyzed using the F test of equality of variances ., Once the data sets were determined to show homoscedasticity , Students t test was performed to test the significance of the differences between the sample conditions ., To verify dose-dependence of cell proliferation rate , ANOVA for regression analysis was used .

Introduction, Results, Discussion, Materials and Methods

In an effort to identify the functional alleles associated with hepatocellular carcinoma ( HCC ) , we investigated 152 genes found in the 4q21-25 region that exhibited loss of heterozygosity ( LOH ) ., A total of 2 , 293 pairs of primers were designed for 1 , 449 exonic and upstream promoter regions to amplify and sequence 76 . 8–114 Mb on human chromosome 4 ., Based on the results from analyzing 12 HCC patients and 12 healthy human controls , we discovered 1 , 574 sequence variations ., Among the 99 variants associated with HCC ( p < 0 . 05 ) , four are from the Dickkopf 2 ( DKK2 ) gene: three in the promoter region ( g . -967A>T , g . -923C>A , and g . -441T>G ) and one in the 5’UTR ( c . 550T>C ) ., To verify the results , we expanded the subject cohort to 47 HCC cases and 88 healthy controls for conducting haplotype analysis ., Eight haplotypes were detected in the non-tumor liver tissue samples , but one major haplotype ( TAGC ) was found in the tumor tissue samples ., Using a reporter assay , this HCC-associated allele registered the lowest level of promoter activity among all the tested haplotype sequences ., Retention of this allele in LOH was associated with reduced DKK2 transcription in the HCC tumor tissues ., In HuH-7 cells , DKK2 functioned in the Wnt/β-catenin signaling pathway , as an antagonist of Wnt3a , in a dose-dependent manner that inhibited Wnt3a-induced cell proliferation ., Taken together , the genotyping and functional findings are consistent with the hypothesis that DKK2 is a tumor suppressor; by selectively retaining a transcriptionally inactive DKK2 allele , the reduction of DKK2 function results in unchecked Wnt/β-catenin signaling , contributing to HCC oncogenesis ., Thus our study reveals a new mechanism through which a tumor suppressor gene in a LOH region loses its function by allelic selection .

Liver cancer is one of the most lethal human cancers ., Identifying functional alleles associated with liver cancer can provide new insights into the disease’s pathogenesis and help to advance the development of new therapeutic approaches ., We conducted re-sequencing of the 4q21-25 region that frequently showed loss of heterozygosity ( LOH ) in liver cancer ., Among the 99 variants associated with liver cancer , four are found within the Dickkopf 2 ( DKK2 ) gene ., We conducted haplotype analysis of the DKK2 promoter sequence and found that a transcriptionally inactive DKK2 allele was selectively retained in the tumor tissues ., Additionally , by sequencing individual molecular clones , we detected 7-mer CCTCCCT sites within the DKK2 promoter region that are involved in PRDM9 binding , pinpointing hotspots for recombination and genome instability ., Furthermore , we demonstrated that DKK2 functioned as an antagonist within the Wnt/β-catenin signaling pathway ., Our findings have led to the discovery of a new mechanism whereby a tumor suppressor gene in a LOH region loses its function by allelic selection .

medicine and health sciences, luciferase, enzymes, gene regulation, population genetics, carcinomas, cancers and neoplasms, enzymology, cell processes, gastrointestinal tumors, liver diseases, oncology, gastroenterology and hepatology, gene types, dna, population biology, promoter regions, cell proliferation, tumor suppressor genes, chromosome biology, proteins, oxidoreductases, gene expression, hepatocellular carcinoma, biochemistry, carcinogenesis, haplotypes, cell biology, nucleic acids, genetics, biology and life sciences, evolutionary biology, suppressor genes, chromosomes

journal.pcbi.1003863

The SH2 Domain Regulates c-Abl Kinase Activation by a Cyclin-Like Mechanism and Remodulation of the Hinge Motion

The expression of the constitutively active BCR-ABL fusion tyrosine kinase is sufficient for the initiation and maintenance of chronic myelogenous leukemia ( CML ) in humans 1 ., BCR-ABL is the result of the t ( 9;22 ) chromosomal translocation that leads to the fusion of the Abelson tyrosine kinase ( ABL1 ) and the breakpoint cluster region ( BCR ) gene 2 , 3 ., The dysregulated fusion protein activates a number of signaling pathways associated with inhibition of apoptosis and uncontrolled proliferation ., In the light of the above it is not surprising that the mechanisms regulating the activation and deactivation of both the wild type c-Abl and BCR-ABL tyrosine kinases have attracted a considerable interest 4–9 ., In physiological conditions the catalytic activity of tyrosine kinases is tightly regulated through the interplay between various protein domains , phosphorylation events and associated conformational states of the catalytic domain ( CD ) 10 ., During the catalytic cycle , its high intrinsic flexibility allows the CD to react to the regulatory elements by switching reversibly between a number of distinct active and inactive states ., In most non receptor-type tyrosine kinases , the catalytic domain is preceded by a Src homology 2 ( SH2 ) domain 11 ( Figure 1A ) ., The importance of the SH2 domain in the auto-inhibition and/or activation of the catalytic domain has been shown in c-Src 6 , 8 , 12–14 , Hck 15–17 , Fes 18 , 19 and c-Abl , among others ., The role of the SH2 domain in c-Abl is of special interest , because it is involved both in auto-inhibition and activation of the CD 18 , 20 , 21 , and mutations in the SH2 domain have been related to imatinib-resistance in CML patients 18 , 19 , 22 ., In the auto-inhibited state , the SH3 and SH2 domains and the SH2-kinase linker form a rigid clamp around the CD , which is locked in place by an N-terminal myristoyl modification of the N-terminal cap region inserted deeply into the CD 7 , 23 ( Figure 1B ) ., This grip reduces the flexibility of the CD and , in particular , dampens the opening and closing of its N- and C-termini around the active site 16 , ., This so-called hinge or breathing motion of the CD is required for catalysis , and its impairment is associated with low catalytic output 26–28 ., While the molecular basis for the role of the SH2 and SH3 domains in Abl autoinhibition is well understood , the mechanism of their activating effect is less straightforward ., Recent crystal structures and small angle X-ray scattering studies have revealed that the transition from the auto-inhibited to the fully activated form of Abl requires a complete reassembly of the complex formed by the CD , the SH2 and the SH3 domains , leading to migration of the SH2 domain from the C-lobe to the N-lobe of the CD ( “top-hat” conformation ) ( Figure 1C ) 24 ., Importantly , the effect of this rearrangement is not reduced to merely revoking the autoinhibition of Abl by removing the SH2-SH3 domain clamp , but the SH2 domain , when bound to the N-lobe of the CD , enhances the activity of the kinase , although it bears no direct contact to the catalytic site ., Recently , it was shown that the I164E mutation in the SH2 domain , which interrupts the hydrophobic interactions at the interface between SH2 and CD in the “top-hat” conformation , leads to deactivation of Abl 5 , 18 ., A similar domain arrangement and activating effect has been observed in other kinases 19 , 29 , such as Fes 18 and Btk 30 ., Hence , in multiple tyrosine kinases , the SH2 domain acts as an allosteric effector ., Comparison of the crystal structures of the auto-inhibited and the activated forms of Abl does not reveal any marked conformational changes , particularly at the active site , that could explain the mechanism of activation by the SH2 domain , a finding that points towards a dynamic rather than static nature of the allosteric effect ., The essential features of this allosteric effect and the molecular mechanism by which it is transferred from the N-lobe to the catalytic site still remain elusive ., The SH3/linker region has also been shown to be involved in the regulation of Abl activation 31 ., However , here we focus on the SH2 domain that , even in the absence of SH3 and the linker , has been shown to have strong activating effect ., We used a multi-disciplinary approach combining elastic network models , extensive molecular dynamics simulations , free energy calculations and functional assays following mutagenesis to characterize the allosteric coupling of the CD of Abl with the SH2 domain as well as the modulation of the dynamic properties of the assembly by interactions in the “top-hat” conformation ., Both the long atomistic simulations and the elastic network models indicate a significant change in the dynamics of key regulatory elements , providing a simple explanation of the mechanism of allosteric activation ., Based on the computational results we designed a number of point mutants to validate the proposed model ., These mutants were expressed in human cells and tested for kinase activity ., We identified mutations , all distant from the active site , that were either activating the catalytic output of the kinase or were disrupting ., Interestingly , we also identified a residue that when mutated lead to a decoupling of the activity of the CD from interaction with the SH2 domain ., Collectively , the data suggested an effect of SH2 binding that results in changes of the properties of the αC helix , reminiscent of the effect that cyclins exert on cyclin-dependent kinases ., To elucidate the mechanism by which the SH2 domain stimulates the catalytic activity of the Abl kinase we first characterized the dynamics of the CD alone and with the SH2 domain bound in the activating conformation using elastic network models ., In the free CD , the two predominant modes emerging from the normal mode analysis ( NMA ) corresponded to the well-described hinge motion 26 , 28 , 32 and to a twist of N- and C-lobe against each other ( Supplemental Figure S1A ) ., When the SH2 domain was included in the elastic network model , the hinge motion continued to be the principal motion , but the corresponding normal mode included a sliding of the SH2 domain along the binding interface , while the amplitude of the movement of the N-lobe of the CD was reduced ( Supplemental Figure S1B ) ., This finding suggests that one role of the SH2 domain may be to regulate the hinge motion while restricting lobe twists during catalysis ., We analyzed the allosteric coupling of local conformational fluctuations along these normal modes 33 ( see Methods and Text S1 ) ., Figure 2A and Supplemental Figure S1C show that local distortions in the SH2 domain are associated with conformational changes in both lobes of the CD ., Important couplings are detected between the SH2 domain and specific , spatially separate motifs in the C-lobe , including the catalytically important activation loop ( A-loop ) , the αD-αE loop , the αF-αG loop and the αG helix ., The αD-αE loop forms part of the myristoyl binding pocket ., The αG helix is known to have an important role in the catalytic mechanisms of many kinases ., In some of them , c-Abl among them , it forms part of a platform for substrate binding 34 ., Furthermore , it has been proposed that in Abl and EGFR the αF helix and the αF-αG loop are allosterically coupled to the αC helix and are involved in the dynamically enhanced stabilization of active conformations 20 , 29 ., Virtually all other motifs coupled to the SH2 domain in turn also couple to the crucial A-loop ., The allosteric coupling analysis thus suggests that the SH2 domain acts primarily on the N-lobe loops and the A-loop ., All these motifs are coupled among them through a dense network of allosteric interactions , which affect also the P-loop and the αC helix , two structural motifs that , together with the A-loop , participate actively in the catalytic process ., We next carried out 1 µs long all-atom molecular dynamics ( MD ) simulations of the free CD and the activated SH2-CD complex in solution ., The structure of the CD does not , in general , deviate much from the crystal structure ( Supplemental Figure S2A , blue line ) ., However , some motifs have a much higher degree of intrinsic flexibility than others , as can be seen from the root mean square fluctuations ( RMSF ) of the backbone ( Figure 2B and Supplemental Figure S2B , blue lines ) ., The largest fluctuations were observed in the N-lobe loops , the αC helix , and the A-loop ., Enhanced flexibility of the N-lobe together with partial unfolding of the αC helix has been observed in other protein kinases , such as FES 18 and EGFR 35–37 ., The flexibility of these motifs are thought to be crucial for the regulation of activity and sensitivity towards specific kinase inhibitors 38–40 ., The partial unfolding of the αC helix results in the rupture of a conserved salt bridge ( Glu305 to Lys290 ) , which needs to be formed and maintained stable for Abl activation 41 ( Figure 3C , blue line ) ., When the SH2 domain was included in the simulation , the CD domain is more rigid ( Figure 2C and Supplemental Figure S2A , orange line ) , while the relative orientation of the two domains and the position of the SH2 domain change noticeably ( Figure 3A and Supplemental Figure S2A , red line ) ., Compared to the crystal structure , conformations in which the SH2 domain is shifted towards the active site predominate ., A further hint at the mechanism by which the SH2 domain is connected to the dynamics of the catalytic domain is given by the comparison of the corresponding flexibility patterns ( Figure 2B , C and Supplemental Figure S2B ) ., By far the strongest impact is on the P-loop , which is known to be vital for ATP binding and kinase selectivity 36 ., Figure 3B compares the most representative structures of the CD in absence ( blue ) and presence ( red ) of the SH2 domain , obtained by a cluster analysis of snapshots from the trajectories ., The SH2 domain directs the β3-αC loop towards the active site and the A-loop ( Supplemental Figure S3A , B ) ., In this conformation , the β3-αC loop is stacked over the P-loop , fixing it in a conformation in which it points towards the DFG motif in the active site and the αC helix ., This interplay between β3-αC loop , P-loop and αC helix does not take place in the free CD ., Furthermore , the rearrangement and stiffening of the N-lobe motifs results in a stabilization of the crucial salt bridge between Glu305 and Lys290 ( Figure 3C , red lines ) ., While the SH2 domain strongly stabilizes the N-lobe in general and in particular the P-loop , it enhances the flexibility of the A-loop and the αF-αG loop ., The A-loop plays a fundamental role in the inactive-to-active conformational switch and together with the other elements , is involved in substrate/product binding ( Supplemental Figures S2C , D ) ., It is thus conceivable that the interaction with SH2 might play a role both in the inactive-to-active equilibrium and in the release of the products , which has been proposed to be the rate determining step of c-Abl-dependent phosphorylation ., Next , we addressed the issue of how the SH2 domain affects global motions of the CD by using principal component analysis ( PCA ) , which provides a description of the dominant motions of the CD during the simulations ., Figure 3D shows the projection of the trajectories in absence and presence of the SH2 domain on the eigenvectors ( EVs ) of the two most predominant modes of motion of the backbone of the CD ., These principal eigenvectors match the first normal modes of the simplified elastic network model quite well , representing again the hinge motion and the lobe-twist , The only significant difference is found in the second PCA eigenvector , that also includes a distortion of the αC helix ( Supplemental Figure 1D ) ., The agreement of these two completely independent methods in describing global changes in the CD dynamics is remarkable and indicates that the conformational changes observed during the PCA are facilitated by the low-frequency , global motions that are intrinsic to the structure ., The SH2 domain significantly restricts the movement along both EVs and leads to changes in the conformational equilibrium ., The amplitude of the hinge motion along EV1 is shifted towards conformations in which the lobes close down over the active site ( Figure 3D , left side of the graphic , and Supplemental Figure S1D , blue lines ) , which is also reflected directly by the distance between N- and C-lobe ( Supplemental Figure S3C ) ., The effect of the SH2 domain on the motion along EV2 is also strong ., The twist between the N- and C-lobes , which leads to conformational changes at the active site of the kinase , is constrained and the distortion of the αC helix is strongly reduced ( Figure 3D ) ., A PCA analysis of the trajectory of the SH2-CD construct revealed that , unsurprisingly , the first eigenvector still represents the pure hinge motion ., However , eigenvectors 2 to 4 all include a certain degree of hinge closing but virtually no distortion of the active site motifs ( Supplementary Figure S1E ) ., The corresponding eigenvalue spectra furthermore confirm that in presence of the SH2 domain the hinge motion ( EV1 ) gains importance relative to all other modes ( Figure 3E ) ., The MD simulations indicate that the SH2 domain bound in the “top-hat” conformation exerts a double effect ., On the one hand , it locally reduces the flexibility of a number of structural motifs of the CD , which participate in the catalytic process , such as the P-loop , the β3-αC loop and the αC helix , locking them in their active conformations ., On the other hand , it modifies the collective motions of the CD , channeling energy away from unproductive twists and distortions into the catalytic hinge motion , and shifting the hinge motion towards closed conformations and the A-loop towards more “active-like” conformations ., To investigate the effects of the SH2 domain on the A-loop conformation propensities we computed the inactive-to-active free energy landscape ., To that aim we used a multiple-replica free energy algorithm ( parallel-tempering Metadynamics ) and a structure based hybrid force field ., A similar computational strategy has been successfully used in the case of the CD of the highly homologous Src kinase to study the A-loop opening , where it was able to provide an accurate reconstruction of the free energy surface underlying the transition 42 ., This force field reproduces fairly well the flexibility patterns observed with long all-atom explicit solvent simulations , apart from a small discrepancy in the region corresponding to the αG-helix in the CD ., This region appears to be somewhat more rigid that it should be in the CD , but recovers the correct flexibility in the complex ( Supplemental Figure S4 ) ., In absence of SH2 , the A-loop of the CD is mostly closed , as expected from its in-vitro low catalytic activity ( Figure 4 , left ) ., The A-loop active-like , or “open” , conformation is still a local minimum of the free energy but at a much higher value ( ΔGO-C≈6 kcal/mol ) ., Moreover , the large free energy barrier separating the closed to the open A-loop state ( ΔG‡-O≈14 kcal/mol ) disfavors the transition ., In contrast , the effect of the SH2 regulatory domain is to shift the equilibrium towards an active-like conformation of the A-loop ( Figure 4 , right ) rendering it as stable as the closed conformation ., The free energy barrier between the two states is greatly reduced ( ∼6 kcal/mol lower than without SH2 ) and the open basin is widened increasing the A-loop flexibility ., Based on these computational results , we proposed a number of point mutations , both in the SH2 and in the catalytic domains , that we expect to modify the dynamic interaction between the domains by interrupting essential interactions or altering the flexibility of important motifs ( Table 1 and Text S1 ) ., We mainly focused on mutations in the β3-αC loop and the hinge region , as in the simulations these motifs were most strongly affected by SH2 binding ., Both motifs are known to be relevant for catalysis , so mutations in these regions can be expected to affect kinase activation in general ., However , based on the simulations , we predicted that for some of them the effect should be different in the free CD and in the SH2-CD construct and therefore strengthen our proposal of the how the SH2 domain interferes with c-Abl dynamics ., The effect of the mutations was characterized by assays of c-Abl kinase activity in vivo and in vitro ., None of the candidate mutations have been described previously as clinically relevant and could in principle have either a moderate activating , neutral , or disruptive effect ., The effect of the mutations on the catalytic activity of c-Abl was assessed both in the context of the SH2-CD module and in the isolated CD , in order to discern whether the mutation generally affects the fold or activation state of the kinase domain , or if it specifically targets the SH2-mediated regulation mechanism ., The mutations were introduced in HA-tagged Abl SH2-CD or CD-only constructs which were transiently expressed in human embryonic kidney 293 ( HEK293 ) cells , and the level of their in vivo activity was assessed by phospho-Y412-Abl and total phosphotyrosine immunoblotting of the crude lysates ., The mutated proteins mostly accumulated to levels comparable to the wild-type constructs , indicating that they did not affect protein stability ., The impact on c-Abl enzymatic activity using an in vitro assay with an optimal peptide substrate was assessed for selected mutations ( Figure 5 ) ., As observed previously 5 , under the conditions of this assay , the wild-type SH2-CD module exhibits at least 2-fold higher activity than the wt CD construct ., Since the peptide substrate contains a single phosphorylation site , the observed difference in activity could be ascribed to the SH2 domain-mediated activation that is independent of phosphotyrosine binding and processive phosphorylation events that may ensue ., The comparison of SH2-CD wt and SH2-CD S173N ( a FLVRES motif mutant which abolishes pTyr binding by SH2 ) using this assay shows no difference in kinase activity , hence pTyr binding contributions can be excluded 5 ., Thus , this system should be well suited to study the effects resulting from SH2-kinase domain interactions ., Some of the tested mutations were neutral , such as N165A and E187K in the SH2 domain ( Supplemental Figure S5 ) , or F330A in the β5 sheet ., Others abolished kinase activity and could not be rescued by the SH2 domain , like K266E in the β1-β2 loop , P328G P329G in the β4-β5 loop , F330A in the β5 sheet , or G340P in the hinge ., In these cases , the effect of the mutation goes beyond interruption of the CD-SH2 interaction and interferes directly with the catalytic mechanism ., All of the tested mutations at the T291 position in the β3-αC loop exerted a moderately disrupting effect both in the context of the SH2-CD module , as well as the CD alone , suggesting that the effect of the mutation is not likely to be explainable solely by the abolishment of the interaction with the SH2 domain ( Supplemental Figure S6 ) ., It is possible that T291 plays an important role due to its localization in the key β3-αC loop , where it might be required to confer or sustain an active conformation of the kinase domain ., A number of mutants ( Figure 1D ) , however , were more revealing in terms of understanding the SH2 activation mechanism ., M297 in the β3-αC loop turned out to be very sensitive to mutations ., Based on the MD simulations , we hypothesized that the β3-αC loop may act as a lever , transmitting the signal from the SH2 domain to the catalytic site and positioning the αC helix correctly ., The relatively conservative M297L mutation led to a several-fold decrease in kinase activity ( Figure 5A ) ., Surprisingly , the more drastic change of the M297G mutation did not impair kinase activity , but had a slightly activating effect on the isolated CD , while it was neutral in the context of the SH2-CD construct ( Figure 5A , B ) ., The slight increase in CD activity had not been anticipated from the simulations of wt c-Abl , and suggests a de novo effect , which underlines the importance of this region for modulating c-Abl activity ., In the presence of SH2 , the activating effect of M297G was either suppressed , masked by similar changes , or compensated for by a corresponding drop in c-Abl activity ., This suggests that SH2 indeed uses the β3-αC region as a key lever to efficiently redirect the conformational changes of CD ., Changing the side chain and therefore the highly sensitive interaction network ( M297L ) severely decreases c-Abl activity , while introduction of additional flexibility ( M297G ) has a slightly activating effect and makes c-Abl activation less dependent on SH2 domain binding ., The pivotal role of the β3-αC loop is further confirmed by the effect of mutating E294 and V299 to prolines , which are the equivalent residues in c-Src , a protein closely related to c-Abl but not known to be activated by SH2 ., We would expect the E294P V299P double mutant to stiffen the β3-αC loop lever and , consequently , the αC helix ., In turn , this should activate the CD and enhance the effect of the SH2 domain ., The double mutant was indeed found to be markedly activating , both in the context of the catalytic domain alone as well as within the SH2-CD construct ( Figure 5C ) ., The introduction of E294P and V299P resulted in a substantial increase in enzyme velocity compared to wt SH2-CD ( Figure 5D ) ., The single mutation E294P also exerted an activating effect on Abl activity , however the effect was more pronounced in the presence of V299P , suggesting that the two mutations might act synergistically ( Figure 5C ) ., Lastly , we have investigated the effect of changes in the flexibility of the hinge region ., In the MD simulations , Y339 in the hinge region showed the highest fluctuations ., Mutation of the tyrosine to glycine , which should make the hinge even more flexible , has no effect on c-Abl kinase activity ( Figure 5E ) ., In contrast , the Y339P mutation , which should rigidify the hinge region , was indeed found to be disruptive to c-Abl activity ( Figure 5E ) ., Mutation of another flexible hinge residue , G340 , to proline also abrogated c-Abl activity , as observed by a decrease in phosphorylation of cellular proteins on tyrosine ., Finally , another striking evidence of the stabilizing effect of the SH2 domain on the kinase domain comes from the changes of in vitro measured c-Abl activity upon temperature increase ( Figure 5F , G ) ., At higher temperature , the activity of the c-Abl CD constructs decreased substantially , most likely due to an increase in mobility that is undirected and unproductive ., However , under the same conditions , the activity of c-Abl SH2-CD increased , suggesting that the SH2 domain is able to direct the enhanced movements of the kinase towards more productive states , as had been indicated by the changes in the PCA eigenvalue spectrum due to SH2 binding ., Interestingly , the Y339P mutation which should rigidify the hinge region , could not be rescued by an increase in temperature , and the activity of the mutant SH2-CD was only slightly raised at 35°C as compared to CD Y339P ( Figure 5F ) ., Contrary to the Y339P mutant in the hinge region , the M297G and E294P V299P mutants in the β3-αC loop preserve the effect of the temperature increase ( Figure 5G ) , suggesting that it is , in fact , the hinge motion that is responsible for the effect ., We suggest that the hinge region always has to maintain an important degree of flexibility in order for the kinase to be functional ., In the free CD , conformations with a large opening of the hinge dominate , and additional twists and distortions render the hinge motion largely ineffective ., Upon SH2 binding , the non-catalytic motions are strongly restricted and the hinge motion is directed to the optimal amplitude and opening needed for catalysis ., In the experiments , the M297G single mutant and the E294P V299P double mutant modified the activity of the CD as well as the effect of the SH2 domain in the activating “top-hat” conformation ., Based on the simulations of the wt kinase , we had predicted that changes in the flexibility of the β3-αC loop should affect the directing effect the SH2 domain exerts on the CD ., We carried out unbiased 1 µs simulations of the free CD and the CD-SH2 construct of both mutants to gain insight at atomic level into the mechanism underlying the observed effects and test our hypothesis ., The M297G mutant , which was chosen to introduce additional flexibility in the β3-αC loop and weaken the coupling of the SH2 domain with the αC helix , was slightly activating in the free CD and neutral in the CD-SH2 construct ., In the simulation of the mutant CD , the expected enhanced flexibility of the β3-αC loop was confirmed ( Supplemental Figure S7A ) ., The simulation also provided us with an explanation for the rather unexpected increase in activity of the CD ., The expansion of accessible conformations lead to the loss of interactions between the β3-αC and the P-loop , and enabled the system to assume a conformation similar to that observed in the wt in the presence of SH2 without requireing the directing effect of the SH2 domain ( Supplemental Figure S8A , C ) ., Contrary to the wild type , the flexibility of the CD in the M297G complex was enhanced , not reduced upon SH2 binding ( Supplemental Figures S7B , S8B ) , and no additional strengthening of the salt bridge between the αC helix and the β3 sheet was observed ( Supplemental Figure S7C ) ., The picture emerging from the simulations is thus that the M297G mutant in the CD partly mimics SH2 binding ., At the same time , due to the enhanced flexibility of the N-lobe loops , the effect SH2 domain on the arrangement of the catalytic important motifs is much smaller than in the wt , leading to a situation , where the difference in activity between CD and SH2-CD is markedly reduced ., To summarize , the simulations indicate that introduction of a glycine residue in the β3-αC loop enhances the flexibility of the N-lobe , allowing the P-loop to adopt an activated conformation , while it partly uncouples kinase activation from SH2 domain binding ., In the E294P V299P double mutant , the introduction of two proline residues in the β3-αC loop was expected to stiffen it , limiting the distortions of the αC helix ., In our experiments , the E294P V299P double mutant was , indeed , found to be significantly activating , while , in contrast to the M297G mutant , the enhancing effect of the SH2 domain was preserved ., In the MD simulation , the mutant CD exhibited a flexibility pattern very similar to the wt with a general reduction in flexibility , which was strongest in the P-loop , the β3-αC loop and the αC helix ( Supplemental Figure S7D and Supplemental Figure S8D ) ., Surprisingly , the SH2 domain , however , increased the conformational fluctuations ( Supplemental Figure S7E ) ., Closer inspection of the corresponding structures , however , showed that this is due to concerted movements of P-loop , β3-αC loop and αC helix , which follow the movement of the SH2 domain and enhance the hinge motion ( Supplemental Figure S8E ) ., The essential salt bridge between E305 and K290 is strongly stabilized even in absence of SH2 ( Supplemental Figure S7F ) ., In principle , the substitution of Glu294 , in close vicinity of the SH2 domain , could alter the domain-domain interplay , above and beyond changes in flexibility , due to the loss of possible electrostatic interactions between the glutamate side chain and residues of the SH2 domain ., We compared the essential dynamics of wt and mutant , and found that in the free CD the hinge movement and the lobe twist are strongly affected by the mutation , while they remain practically unchanged in SH2-CD ( Supplemental Figure S8F , G ) ., This finding supports the view that it is the altered dynamical properties rather than lost interactions between E294 and the SH2 domain that leads to the changes in kinase activation ., The combination of structure-based normal mode analysis , unbiased MD simulations and free energy calculations with experiments in vitro and in vivo has shown how the SH2 domain in the “top-hat” conformation changes the dynamics of the catalytic domain of the Abl kinase ., Based on the MD simulations , we could identify three residues ( M297 , E294 and V299 ) that are key to the interplay between catalytic and SH2 domain in different ways ., Contrary to what could have been expected , the residues forming the hydrophobic spine are not directly affected by the formation of the activating complex ., The picture that emerges from our studies is that the SH2 domain stabilizes and repositions the loops that form the binding site , which , in turn , interact with the surrounding loops , stabilizing the P-loop and the αC helix , redirecting the movement of the N-lobe and changing the inactive to active conformational equilibrium of the A-loop ., The stabilization and repositioning of the αC helix also leads to an enhancement of the salt bridge between E305 and K290 , which has been identified as one of the structural elements essential for kinase activation ., The β3-αC loop preceding the αC helix is the key player in this complex mechanism , acting as a lever that transmits the effect from the SH2 domain-binding site towards the substrate binding residues and the catalytically important motifs ., The effects of the M297G and E294P V299P mutants demonstrate the crucial role of the β3-αC loop ., The observation that these mutations have a different impact on the CD and on the SH2-CD constructs shows that the β3-αC loop is involved in Abl activation by the SH2 domain ., While the introduction of a glycine breaks the chain of transmission from the SH2 domain towards the active site and uncouples SH2 domain binding and kinase activation , the stiffening of the β3-αC lever by the introduction of two proline residues enhances the activating effect of the SH2 domain ., The finding that at higher temperatures the unbound catalytic domain becomes less effective while the activity of the CD-SH2 construct increases indicates that the main role of the SH2 domain is to channel kinetic energy into directed , catalytically relevant motions ., We propose that the SH2 domain modifies the so-called hinge motion , an opening and closing of the N-and C-lobes upon the active site ., This is underlined by the fact that stiffening the hinge by the Y339P mutation decreases substantially the activity of the free CD , which cannot be rescued in the CD-SH2 construct ., The hinge motion and the other changes in dynamics upon SH2 domain binding are involved in both the inactive-to-active conformational changes of the A-loop and the release of the products , the phosphorylated substrate and ADP ., The latter has been proposed to be the rate-determining step of the phospho-transfer reaction in some kinases 38 , 43 ., Analysis of the flexibility pattern of the M297G and E294P V299P mutants reveals that the SH2 domain does not simply stabilize catalytically important motifs but that it regulates the Abl kinase through a complex interplay between stabilization , subtle local conformational changes and direction of the concerted hinge motion ., It is interesting to note that the effect on the β3-αC loop are reminiscent of the binding of cyclin to cyclin-dependent kinases 44 , 45 and the dimerization of the EGF Receptor , in which the catalytic domain of the activator kinase engages the helix αC of the receiver kinase 46 ., The effect of the SH2 domain in Abl , however , occurs at a distance , exploiting a network of interactions and motions that effectively may be equivalent to interaction with the so-called αC “hydrophobic patch” by differ

Introduction, Results, Discussion, Methods

Regulation of the c-Abl ( ABL1 ) tyrosine kinase is important because of its role in cellular signaling , and its relevance in the leukemiogenic counterpart ( BCR-ABL ) ., Both auto-inhibition and full activation of c-Abl are regulated by the interaction of the catalytic domain with the Src Homology 2 ( SH2 ) domain ., The mechanism by which this interaction enhances catalysis is not known ., We combined computational simulations with mutagenesis and functional analysis to find that the SH2 domain conveys both local and global effects on the dynamics of the catalytic domain ., Locally , it regulates the flexibility of the αC helix in a fashion reminiscent of cyclins in cyclin-dependent kinases , reorienting catalytically important motifs ., At a more global level , SH2 binding redirects the hinge motion of the N and C lobes and changes the conformational equilibrium of the activation loop ., The complex network of subtle structural shifts that link the SH2 domain with the activation loop and the active site may be partially conserved with other SH2-domain containing kinases and therefore offer additional parameters for the design of conformation-specific inhibitors .

The Abl kinase is a key player in many crucial cellular processes ., It is also an important anti-cancer drug target , because a mutation leading to the fusion protein Bcr-Abl is the main cause for chronic myeloid leukemia ( CML ) ., Abl inhibitors are currently the only pharmaceutical treatment for CML ., There are two main difficulties associated with the development of kinase inhibitors: the high similarity between active sites of different kinases , which makes selectivity a challenge , and mutations leading to resistance , which make it mandatory to search for alternative drugs ., One important factor controlling Abl is the interplay between the catalytic domain and an SH2 domain ., We used computer simulations to understand how the interactions between the domains modify the dynamic of the kinase and detected both local and global effects ., Based on our computer model , we suggested mutations that should alter the domain-domain interplay ., Consequently , we tested the mutants experimentally and found that they support our hypothesis ., We propose that our findings can be of help for the development of new classes of Abl inhibitors , which would modify the domain-domain interplay instead of interfering directly with the active site .

cell biology, biology and life sciences, computational biology, molecular cell biology, biophysics, molecular biology, biophysical simulations

journal.pcbi.1000575

Computational Model of Membrane Fission Catalyzed by ESCRT-III

Membrane fission leading to division of one continuous membrane into two separate ones is ubiquitous in cell physiology ., It is one of the crucial events in generation of transport intermediates from plasma membranes and intracellular organelles; steady-state dynamics of the endoplasmic reticulum , mitochondria and Golgi complex; virus budding , cytokinesis and other fundamental phenomena ( see for review e . g . 1–3 ) ., In the process of fission , a membrane changes its shape and undergoes a topological transformation which includes transient perturbations of the membrane continuity ., To overcome the membrane resistance to shaping and remodeling , a substantial energy has to be invested into the system , which requires action of specialized proteins ( see for review 2 , 3 ) ., Identification of proteins which shape and remodel membranes in the course of diverse intracellular processes has become a hot topic of cell biology 1 , 3 , 4 ., The major advance has been achieved in discovering proteins generating and/or sensing the membrane curvature ., The list of such proteins is constantly expanding and the mechanisms of their action are being elaborated 1 , 4 , 5 ., Less progress has been made in understanding how proteins drive the membrane fission per se ., While several protein types such as the dynamin-family proteins ( see e . g . 6–10 ) , CtBP1/BARS 11 and PKD 12 have been implicated in fission of cell membranes , until recently , the ability to split membranes was unambiguously demonstrated for , perhaps , only one protein , dynamin-1 9 , 13–15 ., Whereas different versions of the mechanism of membrane fission by dynamin-1 were suggested ( see for review 10 ) , the idea unifying the majority of these proposals is that dynamin self-assembles on the membrane surface into helical oligomers constricting the membrane underneath into thin tubes ., Strong mechanical stresses induced by dynamin in the tubulated membrane upon GTP hydrolysis can relax as a result of membrane division and , therefore , drive membrane fission ., Accumulating evidence suggests that the ESCRT ( Endosmal Sorting Complexes Required for Transport ) complexes 16 – are able to catalyze the membrane budding and fission processes ., The ESCRT machinery consists of five different complexes - theVps27complex ( ESCRT-0 ) , ESCRT-I , -II , and -III , and the Vps4 complex - whose coordinated action sorts trans-membrane proteins into intralumenal vesicles ( ILV ) , which bud off from the limiting membranes of endosomes and transform endosomes into multivesicular bodies ( MVB ) 16–19 ., In addition to the MVB generation , the combined action of ESCRT-III and VPS4 complexes are required for the budding of some enveloped viruses including HIV-1 20and during late steps in cytokinesis 21–24 ., It is thus most likely that ESCRT-III and VPS4 catalyze membrane fission reactions , common to all three biological processes 21–25 ., The ESCRT-III complex in yeast consists of four core subunits Vps20 , Snf7 , Vps24 , and Vps2 26 whose mammalian analogues are the charged multivesicular body proteins CHMP6 , CHMP4 , CHMP3 and CHMP2 , respectively ., The subunits are consecutively recruited to the membrane in the order of Vps20/CHMP6 , Snf7/CHMP4 , Vps24/CHMP3 and Vps2/CHMP2 27–29 and their assembly into higher order complexes was suggested to drive the inward membrane budding in vitro 28 ., Moreover , these four proteins are able to act as minimal budding machinery as was confirmed by demonstration that their sequential addition to giant unilamellar vesicles ( GUV ) generated membrane invagination and abscission of the inward vesicles 29 ., Specifically , formation of membrane buds connected by open necks to the initial membrane was shown to depend , critically , on the Snf7 ( CHMP4 ) and Vps20 ( CHMP6 ) subunits , while the neck fission proved to require the Vps24 ( CHMP3 ) subunits 29 ., Three different albeit similar models for ESCRT-III catalyzed budding have been suggested 30 ., First , Snf7 ( CHMP4 ) circular filaments or flat spirals lying in the membrane plane 31 start at the center of a newly formed membrane bud and catalyze membrane bending as the bud grows 31 ., A second model suggests that a circular ESCRT-III filament with asymmetric ends delineates a membrane patch containing cargo molecules and constricts the neck of an evolving membrane bud via the disassembly action of Vps4 27 ., A third model , similar to the second one , proposes that an ESCRT-III spiral surrounds and constricts a cargo containing membrane domain leading to membrane budding and fission 29 ., However , spiral polymers of ESCRT-III have only been observed for hSnf7 ( CHMP4 ) in vivo 31 and in vitro 32 , whereas the detachment of the forming vesicle including fission of a membrane neck was shown to be crucially dependent on Vps24 ( CHMP3 ) 29 ., Therefore , in addition to the Snf7 ( CHMP4 ) filaments , the structures formed by self-assembly of Vps24 ( CHMP3 ) must play an indispensable role in the ESCRT-III mediated membrane budding and fission ., CHMP3 ( Vps24 ) and CHMP2A ( Vps2 ) form heterodimers 26 , 33 that assemble into tubular nano-structures which display a variety of end-cap shapes including nearly hemispherical dome-like end-caps ( 34 and the section “Experimental support for the model” below ) ., The external and internal radii of these structures are approximately 52 and 43nm , respectively 34 ., In vitro , the AAA ATPase VPS4 binds to the inside of the CHMP2-CHMP3 polymers and leads to their disassembly in the presence of ATP 34 ., The external surface of a CHMP2-CHMP3 nano-structure has a considerable affinity to membranes containing acidic lipids 34 ., Therefore , in the process of self-assembly , the CHMP2-CHMP3 complex must be able to attract a lipid bilayer , hence , scaffolding the bilayer into a strongly curved shape , a process that might drive membrane fission reactions 34 ., In spite of the apparent similarities between the dynamin-I and CHMP2-CHMP3 assemblies such as, ( i ) the ability to scaffold membranes into cylindrical shapes , and, ( ii ) the energy input by nucleotide hydrolysis , CHMPs cannot employ any of the mechanisms of membrane fission suggested for the dynamin action ., Indeed , topologically , the fission reactions mediated by dynamin and ESCRT-III are directed differently: dynamin and its partners drive membrane budding and abscission towards the cytosol , while ESCRT-III mediates membrane abscission away from the cytosol and towards the lumen of an endosome ., Structurally , a membrane portion tubulated by a dynamin oligomer is situated within the protein scaffold and , hence , could undergo further thinning upon detachment from dynamin and divide by self-fusion within the protein framework 14 ., In contrast , the membrane wrapped around a CHMP2-CHMP3 structure is attached to the outside surface of the protein scaffold and , hence , the scaffold hinders the membrane sterically from direct thinning and self-fusion ., Thus , the character of membrane deformation leading to fission driven by CHMP2-CHMP3 structure must differ essentially from that generated by dynamin and the mechanics of the fission reaction must be dissimilar in the two cases ., Here , we suggest and integrate the current structural knowledge on ESCRT-III complexes to elaborate on a novel mechanism of membrane fission by dome-like assemblies formed by the CHMP2-CHMP3 subunits of ESCRT-III ., The essence of our proposal is that , in contrast to the fission mechanisms suggested for the dynamin action ( see for review 10 ) , the site of membrane fission driven by ESCRT-III is not co-localized with the protein scaffold but rather emerges aside of it within a membrane neck which forms in the course of membrane wrapping around the ESCRT-III dome ., The major energy for the fission reaction comes from the energy of membrane attachment to the surface of the ESCRT-III complex ., We discuss a possibility for a reinforcement of the ESCRT-III based mechanism by the Vps4 binding ., Our calculations predict that ESCRT-III domes can serve as effective mediators of membrane fission resulting in generation of vesicles of biologically relevant dimensions ., We consider a hemi-spherical protein dome of radius serving as a scaffold for attachment of a membrane fragment of a total area ( Fig . 2a ) ., While , in reality , the membrane attachment to the dome proceeds concomitantly with the dome assembly , for the calculation purposes we will regard the dome to be completed ., This is based on a plausible assumption that the attractive interaction between the subunits of the CHMP2-CHMP3 structure must be much stronger than all other relevant interactions characterizing the system ., Therefore , the protein self-assembly proceeds irrespectively of the membrane attachment , while the latter follows the dome building and its extent is determined by the interplay between the membrane bending energy and the membrane affinity to the protein surface ., The absolute value of the energy of the membrane interaction with the dome surface per unit area of the membrane-protein interface will be referred to as the membrane affinity and denoted by ., Since the membrane-protein interaction is attractive its energy is negative and its value per unit area is ., Note that , according to our definition , the affinity accounts only for the direct ( probably , electrostatic ) interaction between the protein and the lipid polar groups and does not include the energy of membrane bending , which accompanies the membrane binding to the protein dome and contributes to the total energy of this process ., Therefore , the value of is not supposed to depend on curvature of the protein surface ., In this respect , the notion of the affinity we are using differs from the total energy of the membrane attachment to the protein complex , which includes the bending contribution and is commonly used to characterize interaction of proteins with bent membranes ( see e . g . 1 , 4 , 39 , 40 ) ., In our approach the curvature effects are considered separately from the direct membrane-protein interaction ., The membrane adopts a curved shape of a bud characterized at each point by the total curvature and the Gaussian curvature 41 ., The radius of the narrowest cross-section of the bud neck will be referred to as the neck radius , ( Fig . 2a ) ., The membrane bending energy per unit area of the membrane mid plane , , is given by 42 , 43 , ( 1 ) where is the bilayer bending modulus ( see e . g . 44 ) , and is the bilayer modulus of Gaussian curvature whose values were not directly measured but estimated to be negative ( see e . g . 45 , 46 ) ., We analyze two alternative states of the system: the fore-fission state where the membrane bud is connected by a membrane neck to the membrane portion attached to the protein dome ( Fig . 2a ) , and the post-fission state represented by a separate spherical vesicle and the protein dome completely covered by the membrane ( Fig . 2b ) ., Our goals are, ( i ) to compute the energies of the two states and to find , by their comparison , the affinity values at which the membrane fission event is energetically favorable , and, ( ii ) to determine at which the membrane neck in the fore-fission state becomes as small as guaranteeing fast fission 35 ., In the fore-fission state , the extent of the membrane attachment to the protein dome will be characterized by the angle referred below to as the attachment angle which indicates the position of the upper border of the attached area ( Fig . 2a ) ., The total energy of the system in the fore-fission state , , is the sum of two contributions ., First , the total attachment energy found by integration of the attachment energy density , , over the attached area ., Second , the total bending energy of the membrane , , determined by integration of over the whole area of the membrane including and the area of the bud ., Taking into account Eq . 1 and the system geometry ( Fig . 2a ) , the total energy of the fore-fission state can be expressed as ( 2 ) The first contribution to the Eq ., 2 represents the sum of the attachment energy and the bending energy of the attached membrane portion whose total curvature , , is related to the dome radius , , by ., The second contribution is the bending energy of the bud , which depends on the curvature distribution along the bud surface ., The third contribution is the energy of the Gaussian curvature , which does not depend on the system configuration ., The energy ( Eq ., 2 ) has to be minimized with respect to the attachment angle and the distribution of the total curvature along the surface of the bud for any given value of the affinity ., This will give the equilibrium values for and the corresponding attached area , determine the equilibrium shape of the membrane bud including its neck radius , and provide the equilibrium total energy of the fore-fission state ., Because of a complex shape of the membrane bud , minimization of Eq . 2 will be performed numerically by the standard method of finite elements using the COMSOL Multiphysics software ., In the post-fission state , consisting of a spherical vesicle and the hemi-spherical dome covered completely by the membrane ( Fig . 2b ) the total energy is ( 3 ) In the following , we can skip the Gaussian curvature contribution to the fore-fission energy , and account for the addition of to the energy of the post-fission state ., CHMP2A/CHMP3 polymers were assembled and analyzed by negative staining electron microscopy as described 34 ., CHMP2A/CHMP3 polymers were applied to a holey carbon grid and plunge frozen in liquid ethane ., The samples were examined in an FEI F30 Polara microscope , equipped with a Gatan GIF post-column energy filter 47 ., Tilt series were acquired over an angular range of 120 degrees , at a nominal magnification of 27 , 500 times , which corresponded to a pixel size of 0 . 49nm , and at a defocus of 5 to 7 microns ., Tomograms were generated from these tilt series using the IMOD software package 48 and visualized in Amira ( Visage Imaging ) ., A typical computed shape of the membrane bud corresponding to a certain attachment angle , is presented in Fig . 2a and can be described as a sphere-like cap connected to the attached membrane by a funnel-like neck ., The larger the angle , the smaller the neck radius ( Fig . 3 ) ., At the attachment angle the neck radius becomes smaller than the threshold value , , which fulfills the condition of the fast fission 35 ., Therefore , we limited the considered range of the attachment angles by ., Generally , the computation could be stretched to higher attachment angles corresponding to even narrower necks ., This would require , however , including in the elastic energy model additional terms of higher order in the curvature of the internal monolayer of the neck , and taking into account the energy of the short range hydration repulsion through the neck lumen between the elements of the internal surface of the neck ., Such sophistication of the model would complicate considerably the computation without significant changes of the model predictions on the neck fission ., The character of the dependence of the system energy on the attachment angle is determined by the affinity ( Fig . 4 ) ., According to the first term in Eq . 2 , the membrane binding to the protein dome will occur only if the affinity exceeds a certain value , , which is the least affinity needed for compensation of the energy penalty of membrane bending accompanying the attachment to the dome surface ., At each particular affinity value larger than , the system can reside in a stable or quasi-stable configuration described by the values of corresponding to the energy minima ( Fig . 4 ) ., There are four different ranges of the affinity determining different regimes of the possible system configurations ., Transitions between these regimes are determined by the three characteristic values of the affinity denoted by , and and presented in Fig ., 5 . The first regime corresponds to the affinities smaller than the first characteristic value , ., Here , the energy has one minimum at small values , , of the attachment angle ( Fig . 4 ) , meaning that the stable configuration of the system is a bud with a neck whose radius is somewhat smaller than but comparable with the radius of the protein dome ., We will refer to this configuration as the broad neck configuration ., In the second regime , the affinity varies between the first and the second characteristic values , ., In this range , a second energy minimum emerges at the largest possible attachment angle within the considered range , ( Fig . 4 ) , corresponding to a bud with a neck of radius ( Fig . 3 ) ., This configuration will be called the narrow neck configuration ., The total energy in the second minimum is higher than in the first one , , which means that the narrow neck is a quasi-stable while the broad neck is a stable configuration ., It has to be noted that , in contrast to the first energy minimum , the second one is not characterized by a vanishing first derivative of the energy function and represents the minimal energy value found in the considered range of the attachment angle ., This feature of the second minimum does not influence , however , the conclusions of the analysis of the membrane fission conditions ., In the third regime , the affinity is in the range between the second and third characteristic values , ., Under these conditions , the narrow neck is energetically more favorable ( Fig . 4 ) and , hence , becomes stable whereas the broad neck turns quasi-stable ., Finally , in the fourth regime the affinity is larger than the third characteristic value , ., Here , the energy minimum corresponding to the broad neck vanishes and the only stable state of the system is that of the narrow neck ., The three characteristic affinity values , , and , and the geometrical characteristics of the membrane bud in the four regimes of configurations are illustrated in the phase diagrams ( Fig . 5a , b , c ) ., The first two phase diagrams represents the total energies ( Fig . 5a ) and the corresponding attachment angels ( Fig . 5b ) of the broad and narrow neck configurations for a specific value of the membrane area ., The third phase diagram ( Fig . 5c ) shows how , and depend on the membrane area and , hence , on the area of a vesicle which would form if fission occurs ., All the three characteristic affinities decrease with the membrane area which means that the larger the membrane , the lower affinities are needed for generation of buds with narrow necks ., Recall that we analyze two requirements for membrane fission ., According to the first requirement , the fission reaction has to be energetically favorable meaning that the total system energy in the post-fission state must be lower than in the fore-fission state , ., Upon this condition , the fission reaction may be slow because of the existence of kinetic barriers ., According to the second requirement , the energy barriers of the fission reaction must , practically , vanish , which guarantees fast rates of the membrane splitting ., Particularly , the membrane neck has to narrow up to the threshold value , which guarantees that not just the overall fission reaction but also the intermediate hemi-fission stage is energetically favorable and does not limit the fission rate 35 ., The computed system energies in the fore- and post- fission states for different values of the affinity and different moduli of the Gaussian curvature are presented in Fig ., 6 . According to these results the first requirement is always satisfied in the narrow neck configuration confirming the previous works ., Also for the broad neck configurations the fission reaction may be energetically favorable ., To this end the affinity has to be larger than a certain value varying in the range between 0 . 27mN/m and 0 . 37mN/m for feasible values of the Gaussian curvature modulus ( Fig . 7 ) ., The more negative is , the looser are the fission conditions , i . e . the lower affinity is needed for fission to be energetically favorable ., However , to undergo fission from the broad neck configuration , the system has to overcome a substantial energy barrier and , in practical terms , the membrane splitting will not occur ., The requirement of fast fission can be fulfilled if the system reaches the narrow neck configuration ., However , to achieve this state in the course of the membrane attachment to the protein dome , the system has to proceed through the whole range of the attachment angles beginning from and up to ., This means that the system has to move along one of the energy profiles represented in Fig . 4 ., According to Fig . 4 , if the affinity value is smaller than , there is an energy barrier and the system has to overcome to reach the narrow neck configurations ., This means that for the membrane fission will be restricted kinetically ., At the larger affinity values , , evolution of the membrane bud up to the narrow neck configuration is accompanied by a monotonous decrease of the energy and , hence , proceeds without kinetic restrictions ., Summarizing , the condition for the fast fission is ., To support the model , we studied the structures resulting from the CHMP2-CHMP3 self-assembly by negative staining 34 and cryo electron tomography ( see Materials and Methods ) ., We observed assembly of open tubes , tubes with flat closures , tubes with hemispherical almost closed ends ( defects in closure ) and closed tubular structures with hemi-spherical end-caps ( Fig . 8 ) ., The presence of closure defects observed in the structures assembled in vitro might be due to fact that they have been assembled in the absence of membranes ., In the current model we propose that these structures assemble directly on membranes ., Formation of the closed hemi-spherically capped tubes substantiates the existence of the protein domes which play the central role in the model ., These structures should represent the final stage of CHMP2-CHMP3 polymerization and our model suggests that they are physiologically relevant ., According to our computations , the affinity required to drive fission of the membrane neck depends considerably on the area of the membrane fragment undergoing budding and , hence , on the dimension of the vesicle generated in the result of fission ( Fig . 5c ) ., The ESCRT-III proteins have been implicated in generation of multivesicular bodies ( MVBs ) consisting of vesicles with characteristic diameters between 20 and 100 nm 19 , 53 and in budding of enveloped viruses with diameters varying up to about 100 nm ., Therefore , we performed calculations for the areas of the membrane bud between and corresponding to the relevant range of the vesicle diameters ., The largest affinity denoted as is needed to drive a kinetically unconstructed formation of a bud with a narrow neck of radius less which enables fast fission ., The affinity ( as well as two other characteristic affinities , and , determining conditions for slower fission processes ) , decreases with increasing membrane area ., The maximum value of is needed for generation of the small 20 nm vesicles of MVBs ., According to our results ( Fig . 5c ) , the required affinity is ., The feasible values of the membrane affinity to the protein dome can be estimated based on a thermodynamic analysis of the kinetic measurements of the CHMP2A and CMHP3 monomer binding to the DOPS-SOPC bilayers 34 ., According to these measurements , the CHMP2A and CHMP3 monomers dissociate from lipid with a dissociation rate constant ( koff ) of 0 . 08 s−1 and 0 . 3 s−1 respectively 34 ., The association to lipid for both , CHMP2A and CHMP3 , was found to be diffusion controlled thereby putting a lower limit on the association rate constant ( kon ) of 1×106 M−1 s−1 ., The condition of equilibrium between the lipid-bound and free protein monomers resulting from the equality of the rates of their association to and dissociation from the lipid can be expressed by the equation ( 4 ) where is the number of the lipid-bound protein monomers , is the number of the lipid molecules and is the volume concentration of the free protein monomers ., On the other hand , thermodynamically , the same equilibrium condition can be expressed through the equality of chemical potentials of the lipid-bound and free protein monomers , ( 5 ) where and are the so called standard chemical potentials of the free and lipid-bound protein monomers accounting for the free energy of the direct monomer interaction with the surrounding , and are the contributions of the free and lipid-bound protein monomers from the translational entropy in the solution and on the membrane surface , respectively , is the molar concentration of water molecules ., Eq . 5 takes into account that the whole lipid is organized into one or few extended membranes whose translational entropy has a vanishing effect on the chemical potentials ., The protein-membrane binding energy per protein monomer is related to the standard chemical potentials by , so that the affinity which represents , according to the definition above , an absolute value of the binding energy related to the unit area of the protein-membrane interface , is given by ( 6 ) where is the area of a CHMP monomer exposed to interaction with the membrane ., Combining Eqs . 4–6 we obtain for the affinity ., Given the kinetic constants above , and the estimation for the monomer contact area 33 we determine the membrane affinities of CHMP2A and CHMP3 to be and ., Taking into account that the protein dome consists of the CHMP2A-CHMP3 heterodimers , the average affinity should be about , which exceeds almost by a factor of six the above estimation of for the affinity required for fast fission of the vesicles ., Fission of larger vesicles requires even lesser affinities ., Hence , the binding energy provided by the CHMP-membrane interaction must be excessively large and guarantees fast membrane budding and fission under all biologically relevant conditions ., The suggested mechanism of membrane fission by the ESCRT-III proteins CHMP2A-CHMP3 and the related calculations demonstrate that dome-like assemblies of these proteins could scaffold membrane necks into strongly curved shapes and favor membrane fission ., Since , in contrast to the proteins of the dynamin family , the ESCRT protein complexes attach the membrane to their external surfaces , the fission site emerges within a free membrane fragment aside of the zone of protein-lipid interaction ., The task of the CHMP4 and CHMP6 subunits , which are recruited to the membrane upstream of the CHMP2 and CHMP3 recruitment , is to generate an initial membrane bud with a fixed membrane area whose neck has to undergo fission to complete the vesicle formation ., A role for Vps4 , in addition to its recycling function , can be in reinforcing the wall of the ESCRT-dome which facilitates membrane bending and fission ., It is conceivable that the suggested mechanism is not limited by the action of ESCRT-III proteins but rather has a more general character .

Introduction, Model, Results, Discussion

ESCRT-III proteins catalyze membrane fission during multi vesicular body biogenesis , budding of some enveloped viruses and cell division ., We suggest and analyze a novel mechanism of membrane fission by the mammalian ESCRT-III subunits CHMP2 and CHMP3 ., We propose that the CHMP2-CHMP3 complexes self-assemble into hemi-spherical dome-like structures within the necks of the initial membrane buds generated by CHMP4 filaments ., The dome formation is accompanied by the membrane attachment to the dome surface , which drives narrowing of the membrane neck and accumulation of the elastic stresses leading , ultimately , to the neck fission ., Based on the bending elastic model of lipid bilayers , we determine the degree of the membrane attachment to the dome enabling the neck fission and compute the required values of the protein-membrane binding energy ., We estimate the feasible values of this energy and predict a high efficiency for the CHMP2-CHMP3 complexes in mediating membrane fission ., We support the computational model by electron tomography imaging of CHMP2-CHMP3 assemblies in vitro ., We predict a high efficiency for the CHMP2-CHMP3 complexes in mediating membrane fission .

Membrane fission is a key step of fundamental intracellular processes such as endocytosis , membrane trafficking , cytokinesis and virus budding ., The fission reaction requires substantial energy inputs provided by specialized proteins ., Recently , the ESCRT-III proteins have been implicated in membrane budding and fission involved in multivesicular body formation , cytokinesis and virus budding ., The ESCRT-III proteins self-assemble into circular filaments and flat spirals in the membrane plane and generate tubular structures with dome-like end caps ., We suggest and elaborate computationally on a mechanism by which the ESCRT-III complexes can drive membrane fission ., The essence of the mechanism is in generation in the course of membrane attachment to the dome-like surface of an ESCRT-III assembly of a thin membrane neck accumulating large elastic stresses ., Relaxation of these stresses can drive the neck fission and formation of separate vesicles of biologically relevant sizes ., Estimations of the membrane affinity to the protein surface required for the neck fission to occur and comparison of these values with the experimentally expected values justify quantitatively the proposed mechanism and demonstrate that ESCRT-III assemblies must be highly effective in promoting membrane fission .

biophysics/cell signaling and trafficking structures, biophysics/theory and simulation

journal.pcbi.1002692

Connecting Macroscopic Observables and Microscopic Assembly Events in Amyloid Formation Using Coarse Grained Simulations

A wide range of normally soluble proteins and peptides are known to have a propensity to aggregate into -sheet rich amyloid fibrils ., Such structures do sometimes posses functional roles , for instance as functional coatings and catalytic scaffolds 1 ., However , more often than not , the formation of amyloid structures is a pathogenic event – it is the hallmark of a range of neurodegenerative disorders , including Alzheimers and Parkinsons diseases 2 , 3 ., A diversity of experimental 4–9 , and theoretical 10–15 approaches have been developed to probe the mechanisms of amyloid formation ., Such studies have shed light on the structural and kinetic aspects of amyloid growth; it has , however , proved to be very challenging to characterise the very early stages of this reaction , in particular the primary nucleation events and the subsequent formation of low relative molecular weight oligomers ., Yet there is substantial evidence that these small oligomers , rather than mature fibrils act as neurotoxins , and are implicated in the pathological cascades that underlie neurodegeneration 16–18 ., Hence , despite the technical challenges associated with the study of this phenomenon 19 , 20 , a quantitative understanding of the kinetics of oligomer formation is of great practical and fundamental importance in the context of neurodegeneration and more generally in relation to aberrant protein self-assembly ., The growth of linear aggregates following a nucleation event is described in its simplest form by the classical theory of nucleated polymerisation , developed by Oosawa originally to study the formation of cytoskeletal filaments 21 ., Within this framework , the time dependence of the mass fraction of the fibrils can be expressed as follows 21 , 22: ( 1 ) where is the nucleus size , is an effective rate constant which contains contributions from both the nucleation rate and the elongation rate of the filaments , is the initial mass concentration of the monomers , the mass concentration of the fibrils and the mass fraction of the fibrils ., The key parameters in Oosawas theory are and , since is known a priori ., This formalism has been extended by Ferrone , Eaton and coworkers in their pioneering studies of the aggregation of sickle haemoglobin to include secondary nucleation events such as filament fragmentation and surface catalysed nucleation 23 , 24; in the present paper we focus on homogeneous nucleation which is crucial in the formation of the primary nuclei ., Several groups have reported numerical simulations of the nucleation and growth of amyloid fibrils 25–35 ., Such simulations provide invaluable microscopic insight into the mechanism of amyloid formation ., The focus of the present paper is different: we wish to compute quantities that are directly accessible to experiment , such as the time dependence of fibril formation , and apply to these quantities the same analysis that is applied to experimental data ., This approach allows us to test whether a reliable estimate of the critical nucleus size can be obtained by fitting the experimental fibril-growth curve to an analytical approximation , and sheds light on the microscopic and mechanistic interpretation of the nucleus size ., In addition , our simulations allow us to follow in detail the pathway by which amyloid fibril nucleation takes place and shed light on the condensation-reorganisation mechanism that underlies the primary nucleation process 36 ., Since primary nucleation processes are very rare events , and the critical nucleus is by definition the species after the highest free energy barrier and therefore lowest relative concentration within the aggregation pathway , simulations are currently the most fruitful avenue to access the structural determinants of the critical nucleus and to follow its formation and the subsequent conversion to amyloid fibrils ., In the present paper we study fibril nucleation by considering a highly simplified model of amyloidogenic peptides that captures the salient features of this self-assembly process ., A key property of amyloidogenic peptides is that they change conformation when converting from their normal soluble form into the amyloid state 3 ., Our model has two internal states: one ( denoted the -state ) has weak intra-peptide interactions and represents the soluble random coil structure , the other ( the -state ) has a significantly higher intrinsic internal energy , but has stronger interpeptide attractions ., The higher internal energy of the -state results from the loss of conformational degrees of freedom relative to the -state , and its higher propensity to form inter-peptide contacts is a consequence of the availability of the residues for hydrogen bonding with neighboring peptides in the sheet conformation ., In this picture , fibril nucleation takes place once the free-energy gain due to aggregation compensates the free energy cost of converting the monomers to the -state ., The model that we use is based on a peptide model reported in ref 37 ., that has been extended to account for the two-state nature of amyloidogenic peptides ( see section methods ) ., Because this model is simple and therefore computationally highly tractable , it allows us to study the behavior of large numbers of peptides ., Specifically , we can use it to compute the fibrillar growth profile and the free-energy landscape for oligomer formation ., We used dynamic Monte Carlo simulations to generate trajectories for an ensemble of peptides in a system with periodic boundary conditions ., In what follows , we relate the properties of the present model system to solutions of A peptides with a length of 40–42 amino acid residues , which are the major components of the aberrant deposits found in connection with Alzheimers disease 2 , 3 ., With this mapping , the conditions of our simulations correspond to a concentration range of 0 . 2–8 mM ., For more details on the model and the simulation method , see section methods ., We used Dynamic Monte Carlo ( DMC ) 38–41 for our simulations ., In this method , the displacements , that can occur are so small , that no unphysical moves can occur ., Moreover , the parameters of individual moves ( translation and rotation ) are fitted to experimental timescales such as diffusion and rotational constants ., The values were taken from one of the most studied amyloid forming peptides A and were averaged between its 40 and 42 amino acid long forms ( and at 300 K ) 42 ., Hence , maximum displacement was d nm and maximum rotation and consequently the time of our simulation step ( when on average all particles move , i . e . , sweep ) can be roughly related to 0 . 02 ns ., The amyloidogenic peptides were modeled as Patchy Spherocylinders ( PSC ) 37 , i . e . cylinders with hemispherical caps at both ends and with an attractive stripe on its side ., As was shown in ref ., 37 such particles can either occur in an oligomeric form or assemble into amyloid-like structures with two filaments , depending on the model parameters ., Unlike the model described in ref ., 37 , the present model peptides can occur in two possible states: the first one ( denoted as the -state ) corresponds to the random coil conformation of peptide in solution 43; the second state ( called -state ) corresponds to the -sheet structure found in the fibrils ., The free energy difference corresponding to change from the to the -state , is denoted by ., In what follows , we chose ., These values were chosen to reflect the fact that , in experiments , amyloidogenic proteins are typically not found at detectable concentrations in the -sheet conformation in solution 44 , 45 ., The attractive stripe is responsible for self-assembly and the cross-section of small oligomers in ideal conformation are depicted in Figure 1 ., Chirality was introduced into the model in order to reproduce the relatively long persistence length of the amyloid fibrils ., This was achieved by rotating the attractive patch off the cylinder axis around the vector connecting the middle of the cylinder axis with the middle of the patch . The attractive stripe was thus misaligned with the body of the spherocylinder represented by repulsive potential ., The aspect ratio of the PSC was chosen to mimic the elementary -sheet unit of A peptides with dimensions ., We use an implicit-solvent model where the interaction between the attractive stripes ( patches ) on different peptides effectively includes all possible interactions such as hydrophobic interaction , hydrogen bonds , salt-bridges , etc ., The potential minimum of two interacting -states was −21 with an attractive stripe size of running length-wise in order to mimic the interaction potential of A peptides 11 and being able to form cross beta fibrils ., We considered both a chiral and a non-chiral version of the model ., The -states were interacting with a minimum of −8 . 4 and had a patch size of , which was inspired by the hydrophobic patch of the random coil structure covering 25% of its surface 46 and its presence mainly in monomeric form in solution ., The interaction between and state was calculated using Berthelots rule 47 ., The interaction between the particles is effective , i . e . taking into account all the interactions ., The probability of a PSC to attempt to switch its conformation from the -state to the -state or vice versa was per particle move ., The value was estimated based on the rearrangement time of a polypeptide with a size of 18 Kuhn lengths 5 ., The fibrillar growth with switching probabilities one order of magnitude larger or smaller was without any significant difference ., In all our simulations we employed an NVT ensemble and periodic boundary conditions ., The systems contained 600 PSC with the box sizes ( in nm ) of 50 , 75 , 100 , 125 , 150 and 175 corresponding to concentrations ( in mM ) of 7 . 97 , 2 . 36 , 1 . 00 , 0 . 51 , 0 . 30 , and 0 . 19 ., For each concentration at least three separate runs were conducted with different random initial configurations and the obtained growth profiles were averaged over all runs ., The size of a fibril for the relative mass profiles was defined as all oligomers with a size of at least four monomers ., A representative snapshot of the late stage of a simulation of fibril growth is shown in Figure 2: it reveals that aggregation results in fibrillar species with a morphology similar to that observed in experiments 48 ., Our simulations allow us to follow the time dependence of the aggregation number and hence of the mass of individual aggregates as they grow ., Figure 3 shows a representative time trace ., Initially the system is in a purely monomeric state ., As a result of the collision of two peptides , a dimer can be formed ., The dimer can either dissociate into monomers , or can grow to a trimer through monomer addition ., The oligomers with an aggregation number below four are highly dynamic and interconvert readily between different aggregation states , including dissociation to monomer ., However , tetramers , once formed , always develop into a fiber , which suggests that the size of the critical nucleus is 4 or just below ., In what follows , every oligomer containing four or more monomers is counted as a fiber ., At later times , as larger aggregates emerge , the numbers of monomers and oligomers ( dimers and trimers ) decreases through the incorporation of peptides into larger structures ., Figure 4 shows the average increase in the aggregate mass concentration obtained at a fixed concentration ( 0 . 51 mM ) ., We first tested whether these data could be fitted to the Oosawa theory ., We find that , as is the case for experimental studies 15 , analysis of the system under a single set of conditions does not yield strong enough constraints to allow a reliable estimate of the parameters in Oosawas theory to be obtained ., Indeed , fits of similar quality were obtained with critical nucleus sizes varying between 2 . 0 and 5 . 0 ., This finding therefore highlights the difficulty in resolving microscopic parameters , in this case the critical nucleus size , from macroscopic bulk data at a single concentration since a similar shape of the curve can be obtained for different combinations of the nucleus size and the nucleation rate ., These two parameters can be disentangled , however , when data at different concentrations are considered ., The average growth profiles of the simulation with different initial monomer concentrations were simultaneously fitted to Eq ., 1 ( see Figure 5 ) ., The best fit was obtained for a critical nucleus size =\u200a3 . 8 and a growth rate of , where is the unit time of our simulation roughly corresponding to 0 . 02 ns ( see Methods ) ., We note that , in this formulation , corresponds to the first species in the aggregation pathway that has a higher than 50% probability to grow into a fibril ., The results of the simulation also allow us to test the connections between the characteristic scaling behaviour 15 , 22 , 23 , 49 of the half-time and the critical nucleus size that follows from Eq ., 1 given as the powerlaw: ( 2 ) where is the standard concentration and is the half-time of an aggregation reaction at this concentration ., Following this procedure we obtain a slightly smaller estimate for the critical nucleus size: 3 . 4 ., To test the sensitivity of Oosawas theory in identifying processes other than nucleation and growth , we also employed a model characterized by the absence of chirality compared to the model used previously ( see Methods ) ., In such a system , the persistence length of the polymers is low for bending mode parallel to the narrow dimension of the rectangular cross-section ., In Figure 6A we show that there is a systematic deviation from the fit according to Oosawas theory ., In particular , the growth is always faster at the beginning and slower at the end of the growth than the fit ., The reason is depicted in Figure 6B , where we can see that a flexible fiber can bent into a ring like structure , thereby effectively removing a potential growing site from the system ( i . e . , the fibrillar end is not accessible for further monomer addition ) ., The fusion of fibrils , which we observed for both the non-chiral and the chiral models , also leads to a decrease in the late-stage rate of growth of fibrils ., Neither ring formation nor fusion are accounted for in Oosawas theory , and this fact could explain the slight deviation of the simulation data from the fit shown in Figure 6 ., Note that at lower concentrations not all the monomers are depleted from the bulk at the end of simulation ., This is also the case in the chiral model and reflects the finite probability for monomers to dissociate from the aggregates leading to an equilibrium between the monomeric and aggregated forms of the peptides as is observed in experiments 50 ., The simulations allow us to follow the time evolution of single aggregates as they form ., This information makes it possible to relate the macroscopic average quantities such as the scaling exponents characterising the lag-time , to microscopic events taking place on the level of individual peptides ., Based on classical nucleation theory , the free energy profile along the reaction coordinate ( aggregation number ) increases from the monomer up to a maximum , after which it monotonically decreases under supersaturated conditions ., The point of the maximum free energy is related to the size of the critical nucleus ., We define the nucleus as the first oligomer after this free energy barrier , in other words the nucleus is the smallest oligomer , which has a higher probability to grow than to shrink ., In order to look at the nucleation from yet another perspective , we constructed a free energy landscape for the different kinds of oligomers up to tetramer using the following procedure: The free energy of a mixed oligomer can be decomposed into several parts: ( 3 ) The enthalpic contribution , , to the free energy of an oligomers ( i-mer , where ) was determined as the minimum of the interaction energy of a given oligomer , which is schematically depicted with the enthalpic values in Figure 1 ., Naturally , the bigger the oligomer and the larger the patch , the stronger the interaction is ., Importantly , due to the patch size , a tetramer of -particles ( ) has no enthalpy gain compared to its trimer counterpart ( ) ., The next contribution is the free energy associated with changes in the internal degrees of freedom of the protein molecule , ; this value is large compared to since the experimental measurements only report evidence for the state in solution: no free molecules in the state are observed ., The free energy difference is fixed at 15 for each monomer which changes its state from the soluble state to the state that it assumes in the aggregates ( for more details see methods ) ., The last contribution is entropic , where is the temperature and is the entropy; this contribution stems mainly from the loss of translational entropy upon binding to the cluster , but it also includes the rotational and internal entropy ., Two additional simulations were carried out , each with fixed type of achiral monomers ( pure and pure ) ., The free energy of the i-mers was determined based on their relative populations ( see supplementary information Text S1 ) ., The enthalpic contributions can be determined directly by computing the relevant interaction terms in an ideal configuration ., By subtracting it from the free energy and by dividing by we obtain the entropy per monomer for each ( pure ) i-mer ., Assuming that the entropy of a single particle in a given state is similar in all ( mixed ) i-mers independent of their composition , we can construct the free energy landscape of the oligomers ; again using the enthalpic contribution for an ideal configuration ( see Figure 7 and the supplementary information Text S1 for a detailed description of how to construct it ) ., Note , that the achiral monomers were employed for this calculation as it easier to determine the ideal maximum interaction enthalpy and therefore enthalpic contribution to the free energy ., The nucleation for achiral and chiral model is very similar as they differ mainly in the later stage as self-assembled fibrils in their rigidity ., The analysis of the energy landscape reveals the microscopic details through which primary nucleation occurs in the coarse grained system ., The path of the lowest energy connecting the monomeric states with the aggregates starts with the assembly of a dimer of molecules in an unstructured state ., At this stage , the free energy cost for the conversion of one of the molecules in the dimer to the state is 9 . 2 , less than the cost of the conversion in solution , 15 , but still sufficiently high for this mechanism not to be the major contribution to the overall production of aggregates ., However , the unstructured trimer , obtained through monomer addition from the dimer , possesses a lower free energy than the mixed dimer , and at this stage the conversion of one of the molecules to the state is associated with a significantly lower cost of 3 . 2 resulting in the species ., Subsequent conversions to the state are associated with an even smaller energy cost , and the species is only 0 . 3 higher in free energy relative to the species ., This mixed aggregate represents the species with the highest free energy on the aggregation pathway; subsequent additions of monomers to this nucleus lower its free energy and result in the formation of -sheet fibrils ., Therefore the nucleus size in the formulation of Eq ., 1 is 4 , in excellent agreement with the analysis performed on the average kinetic data ., The first fully -sheet aggregate is the tetramer ., The overall nucleation pathway that emerges from our simulations is that of a nucleation process followed by a conformational conversion ., The conversion step is , however , dependent on the addition of monomers and does not take the form of one step cooperative conversion 36 ., In particular the critical nucleus is a mixed , partially converted , aggregate rather than the fully unconverted species ., This finding is likely to be of general value since the mechanism identified in the present study allows the conversion of a number of monomers in discrete steps combined with monomer addition steps to avoid the high free energy barrier associated with a conversion of the entire aggregate in a single step ., These results therefore generalise the nucleated conformational conversion model 36 to include multi-step conversion ., We have devised a simulation scheme which allows us to study a system of aggregating peptides and compute quantities that are directly accessible in experiments , such as the scaling behaviour of the lag-time with aggregate concentration , and to relate these characteristics to the microscopic events that underlie the generation of single nuclei and their subsequent growth ., In order to reach the long time scales required for such a study , we have employed a coarse-grained model , which includes a representation of the internal degrees of freedom of the polypeptide chain as well as the possibility to assemble into both oligomers and elongated fibrils ., The obtained sigmoidal growth of the oligomers is in agreement with previous studies employing different two state models 25 , 29 ., The nucleus size was found to be in mutual agreement from all the employed methods ., The most trivial one is the empirical observation of the oligomers time distribution , where our results show that all tetramers develop into a fibril ., The second method is the fit of the growth profile to Oosawas theory , which has to be performed with data at varying initial monomer concentration for unambiguous results ., The fitted nucleus size is 3 . 8 ., The same result was obtained from our analysis of the free energy landscape for the oligomer formation ., The coarse grained system that we study possess parameters that are representative of experimental systems , where nucleus sizes of the order of 2–4 are commonly reported 6 , 15 , 51 ., We found very good agreement of our simulated fibril growth with the theory of Oosawa , especially for the chiral model ., The small deviations can be due to the fact that our simulation time is rather small ( somewhere around 2 s , compared to hours in many experiments ) , a factor which required high concentrations in order to observe fibril growth ., Some processes such as fibril fusion could , therefore , be enhanced under the conditions used in the simulations ., Note that the current implementation of our model does not lead to any secondary nucleation 22 , 23 , 51 and the scale of our simulations prevents the fibril breakage 15 , 52 ., As a result we have not observed secondary nucleation pathways which would lead to different kinetics of fibril formation 15 , 22 ., The ability to compute the free energy landscape of the oligomers allowed us to study the sequence of events that leads to the generation of a fully -sheet aggregate from the monomeric precursor state ., We found that the that this conversion occurs concomitantly with the growth of the oligomers through monomer addition ., Furthermore , the critical nucleus is found to be a mixed species including both converted and unconverted peptides , and its size is in good agreement with that determined from the analysis of the scaling behaviour of the average lag-time ., This concerted mechanism allows the energy penalty from the conversion of individual peptides to be compensated by the energy gain from an increase in the number of favorable inter-peptide contacts .

Introduction, Methods, Results, Discussion

The pre-fibrillar stages of amyloid formation have been implicated in cellular toxicity , but have proved to be challenging to study directly in experiments and simulations ., Rational strategies to suppress the formation of toxic amyloid oligomers require a better understanding of the mechanisms by which they are generated ., We report Dynamical Monte Carlo simulations that allow us to study the early stages of amyloid formation ., We use a generic , coarse-grained model of an amyloidogenic peptide that has two internal states: the first one representing the soluble random coil structure and the second one the -sheet conformation ., We find that this system exhibits a propensity towards fibrillar self-assembly following the formation of a critical nucleus ., Our calculations establish connections between the early nucleation events and the kinetic information available in the later stages of the aggregation process that are commonly probed in experiments ., We analyze the kinetic behaviour in our simulations within the framework of the theory of classical nucleated polymerisation , and are able to connect the structural events at the early stages in amyloid growth with the resulting macroscopic observables such as the effective nucleus size ., Furthermore , the free-energy landscapes that emerge from these simulations allow us to identify pertinent properties of the monomeric state that could be targeted to suppress oligomer formation .

A number of normally soluble proteins can form amyloid structures in a process associated with neurodegenerative diseases such as Alzheimers and Parkinsons diseases ., Mature amyloid structures consist of large fibrils containing thousands of individual proteins aggregated into linear nanostructures; there is increasing evidence , however , that the toxic species responsible for neurodegeneration are not the mature fibrils themselves but rather lower molecular weight precursors commonly known as amyloid oligomers ., Unfortunately , these early oligomers are commonly thermodynamically unstable and of nanometer scale dimensions , factors which make them highly challenging to probe in detail in experiments ., We have used computer simulations of a model inspired by Alzheimers Abeta peptide to investigate the early stages of protein aggregation ., The results that we obtain were shown to fit Oosawas polymerization theory , a finding which allows us to provide a connection between the microscopic molecular parameters and macroscopic growth ., One crucial parameter is size of the nucleus , i . e . the basic oligomer existing at origin of the formation of each fiber ., We have revealed a path for the formation of this nucleus and validate its size by several methods ., Our results provide fundamental information for influencing the early stages of amyloid formation in a rational manner .

molecular mechanics, macromolecular assemblies, chemistry, biology, computational chemistry, biophysics simulations, biophysics

journal.ppat.1002899

Establishment of a Reverse Genetics System for Studying Human Bocavirus in Human Airway Epithelia

Human bocavirus 1 ( HBoV1 ) was initially identified in 2005 , in nasopharyngeal aspirates of patients with acute respiratory-tract infections ( ARTI ) 1 ., It was found to be associated with ARTI in children , at a detection rate of 2–19% 2–5 ., Three additional human bocaviruses , HBoV2 , 3 and 4 , discovered in human stool samples , have since been phylogenetically and serologically characterized 6–9 ., However , whether these are associated with any diseases is currently unknown ., HBoV1 is commonly detected in association with other respiratory viruses , and is the fourth most common respiratory virus ( after respiratory syncytial virus ( RSV ) , adenovirus and rhinovirus ) in infants less than 2 years of age who are hospitalized for the treatment of acute wheezing 2 , 10–12 ., Indeed , ARTI is one of the leading causes of hospitalization of young children in developed countries 13 , 14 ., Acute HBoV1 infection , diagnosed by a virus load of >104 genome copies ( gc ) /ml in respiratory samples , viraemia , or by detection of HBoV1-specific IgM or of an increase in the levels of IgG antibodies , results in respiratory illness 2 , 15–20 ., Recent descriptions of life-threatening HBoV1 infections in pediatric patients in association with high virus loads or diagnostic HBoV1-specific antibodies 21–23 , in addition to a recent longitudinal study of children from infants to puberty , documenting a clear association of acute primary HBoV1 infection with respiratory symptoms 24 , strongly support that HBoV1 is an etiological agent of both upper and lower ARTI ., HBoV1 has been classified as a new member of the genus Bocavirus of the family Parvoviridae 25 , of which bovine parvovirus ( BPV1 ) and minute virus of canines ( MVC ) are the prototypes 26 , 27 ., In comparison with the BPV1 and MVC genomes , the HBoV1 genome sequences obtained previously appeared to exclude the two termini , and therefore , were incomplete 28 ., However , sequencing of the head-to-tail junctions of HBoV1 and HBoV3 “episomes , ” which had been amplified in DNA samples extracted from HBoV1-infected differentiated human epithelial cells and from intestinal biopsies of HBoV3-infected patients , respectively , revealed portions of the HBoV termini 29 , 30 ., Notably , these sequences were conserved with the terminal sequences of BPV1 and MVC 28 ., In vitro HBoV1 infection has been reported only once in well-differentiated human airway epithelia ( HAE ) 31 ., That study provided only minimal information on virus replication , and did not include observations of pathophysiology ., Obviously , the lack of a sustainable and highly reproducible system that enables high-yield virus production , as well as the ability to conduct reverse genetics is a significant barrier to further elucidation of HBoV1 replication and pathogenesis ., In the current study , we have successfully sequenced the full-length HBoV1 genome and cloned it in a plasmid referred to as pIHBoV1 ., Furthermore , we have demonstrated that transfection of human embryonic kidney 293 ( HEK293 ) cells with pIHBoV1 results in efficient production of HBoV1 virions at a high titer , and that these virions are able to productively infect both primary and conditionally transformed polarized HAE ., A head-to-tail junction of an HBoV1 episome identified in an HBoV1-infected HAE 28 , 29 was found to possess two sequences ( 3′-CGCGCGTA-5′ and 3′-GATTAG-5′ ) identical to parts of the BPV1 left-end hairpin ( LEH ) 27 , 32 ., This finding suggested that the head sequence is part of the HBoV1 LEH ( nucleotides in blue; Figure 1A ) ., We therefore used the head sequence as the 3′ end of a reverse primer ( RHBoV1_LEH ) ., Together with a forward primer ( FHBoV1_nt1 ) , which anchors the 3′ end of the HBoV1 genome predicted from the BPV1 LEH , we amplified the hairpin of the LEH from a viral DNA extract ( 1 . 2×108 gc/ml ) prepared from a nasopharyngeal aspirate taken from an HBoV1-infected patient ( HBoV1 Salvador1 isolate ) 17 ., Only one specific DNA band was detected at approximately ( ∼ ) 150-bp ( Figure 1D , lane 1 ) ., Sequencing of this DNA revealed a novel sequence of the HBoV1 LEH ( nucleotides in red between the two arrows; Figures 1A and S1A ) ., Because the LEHs of the prototype bocaviruses BPV1 and MVC are asymmetric 27 , 32 , we set up another PCR reaction with a forward primer located in the hairpin ( FHBoV1_LEH ) and a reverse primer targeting a sequence downstream of the LEH at nt 576 ( RHBoV1_nt576; Figure 1B ) ., Sequencing of a DNA fragment ( Figure S1B ) , detected as expected as a ∼600-bp band ( Figure 1D , lane 3 ) , confirmed the presence of the novel joint sequence and the LEH ( Figure 1B ) ., The tail of the HBoV1 head-to-tail junction 28 , 29 was found to contain a sequence ( 5′-GCG CCT TAG TTA TAT ATA ACA T-3′ ) identical to that of the right-end hairpin ( REH ) of the other prototypic bocavirus MVC 27 ., We thus speculated that the entire HBoV1 REH is similar in structure to its MVC counterpart ., Using a reverse primer targeted to this sequence ( RHBoV1_nt5464 ) and a forward primer located upstream of the REH ( FHBoV1_nt5201 ) , we were able to amplify a specific ∼300-bp-long DNA fragment ( Figure 1D , lane 5 ) ., Sequencing confirmed the presence of the palindromic hairpin of the predicted REH ( nucleotides in red; Figures 1C and S1C ) , and revealed two novel nucleotides at the end of the hairpin ( GC in red; Figure 1C ) ., These results indicate that we have identified , for the first time , both the LEH and REH of the HBoV1 genome from a clinical specimen , and confirm that the HBoV1 genome structure is typical of the genus Bocavirus ., We also cloned and sequenced the non-structural ( NS ) and capsid ( VP ) protein-coding ( NSVP ) genes of the HBoV1 Salvador1 isolate from the patient-extracted viral DNA ., We then ligated the LEH , NSVP genes and REH into pBBSmaI using strategies diagramed in Figure S2 , and refer to this full-length clone as pIHBoV1 ., We have deposited the sequence of the full-length genome of the isolate in GenBank ( JQ923422 ) ., As we previously showed that HEK293 cells support replication of the DNA of an autonomous human parvovirus ( B19V ) in the presence of adenovirus helper genes or adenovirus 33 , we first investigated whether the adenovirus helper function is necessary for pIHBoV1 replication in HEK293 cells ., Specifically , we transfected pIHBoV1 into HEK293 cells ( untreated or infected with adenovirus ) , alone or with pHelper ., Interestingly , we found that pIHBoV1 replicated well in the absence of helper virus ., Indeed , all the three representative forms of replicated bocavirus DNA 27 , 34 ( DpnI digestion-resistant dRF DNA , mRF DNA and ssDNA ) were detected in each test case , and at similar levels ( Figure 2A ) ., DpnI digestion-resistant DNA bands are newly replicated DNA in cells as DpnI digestion only cleaves plasmid DNA prepared from prokaryotic cells , which is methylated at the dam site 35 ., In contrast , these DNA forms of the viral genome were absent in pIHBoV1-transfected primary airway epithelial cells ( NHBE; Figure 2B , lanes 7&8 ) and present at very low levels ( over 20 times lower than in pIHBoV1-transfected HEK293 cells ) in pIHBoV1-transfected human airway epithelial cell lines BEAS-2B ( Figure 2B , lanes 5&6 ) , A549 and 16HBE14o- ( Figure 2C ) , even in the presence of adenovirus ., Thus , replication in these cells appears to be non-existent or poor in these contexts ., To confirm the specificity of DNA replication and the identity of the DpnI-resistant DNA bands , we disrupted the ORFs encoding viral proteins NS1 , NP1 , VP1 and VP2 in pIHBoV1; knockout of expression of the corresponding viral protein was confirmed by Western blot analysis ., When the NS1 ORF was disrupted , no DpnI digestion-resistant DNA was detected ( Figure 2D , lane 4 ) , confirming that replication of this DNA requires NS1 ., Notably , when the NP1 ORF was disrupted , an RF DNA band was detected but it was very weak ( Figure 2D , lane 6 ) , suggesting that NP1 is also involved ., When the VP2 ORF was knocked out , the ssDNA band disappeared , but this was not the case when VP1 was disrupted ( VP2 was still expressed; Figure 2D , compare lanes 7 to 9 ) , these findings are consistent with a role for the capsid formation in packaging of the parvoviral ssDNA genome 36–38 ., The presence of the ssDNA band in pIHBoV1-transfected HEK293 cells suggested that progeny virions were produced ., To prove this , we carried out large-scale pIHBoV1 transfection and CsCl equilibrium centrifugation to purify the virus that was produced ., We fractionated the CsCl gradient , and found the highest HBoV1 gc ( 1–5×108 gc/µl ) at a density of 1 . 40 mg/ml , which is typical of the parvovirus virion ., Electron microscopy analysis revealed that purified virus displayed a typical icosahedral structure , with a diameter of ∼26 nm ( Figure 2E ) ., Collectively , these findings confirm that we have generated a full-length clone of HBoV1 capable of replicating and producing progeny virus in transfected HEK293 cells ., The infectivity of the HBoV1 virions purified from pIHBoV1-transfected HEK293 cells was examined in polarized primary HAE , the in vitro culture model known to be permissive to HBoV1 infection 31 ., Three sets ( different donors , culture lots #B29-11 , B31-11 and B33-11 ) of B-HAE were generated , and these were infected with HBoV1 from the apical side ., Initially the B-HAE cultures were infected with various amounts of virus , and when a multiplicity of infection ( MOI ) of ∼750 gc/cell was used , most of the cells ( ∼80% ) were positive for anti-NS1 staining ( indicating that the viral genome had replicated and that genes encoded by it had been expressed ) at 5 days post-infection ( p . i . ) ., This MOI was subsequently used for apical infection ., Notably , B29-11 , B31-11 and B33-11 HAE each supported productive HBoV1 infection ( Figures 3 and S3 ) ., Immunofluorescence ( IF ) analysis of infected B31-11 HAE at 12 days p . i . showed that virtually all the cells expressed NS1 and NP1 ( Figures 3A and 3B ) , and that a good portion of the infected cells expressed capsid proteins ( VP1/2; Figure 3C ) ., The production of progeny virus following HBoV1 infection was monitored daily by collecting samples from both the apical and basolateral chambers of the HAE culture and carrying out HBoV1-specific quantitative PCR ( qPCR; Figures 4A and S3B ) ., In the case of B33-11 B-HAE , apical release was obviously initiated at 3 days p . i . , then continued to increase to a peak of ∼108 gc/µl at 5–7 days p . i . , then decreased slightly through day 10 p . i . and was maintained at a level of ∼107 gc/µl through day 22 p . i . ( Figure 4A ) ., The total virus yield from one Millicell insert of 0 . 6 cm2 over a 24-h interval was greater than 2×1010 gc ., This result suggested that productive HBoV1 infection of primary B-HAE is persistent ., Notably , in the B-HAE cultures from both donors , virus was also continuously released from the basolateral side , keeping pace with apical secretion throughout , though at levels about one log lower than the release from the apical surface ( Figures 4A and S3B ) ., The genomes of the progeny virions released from infected B-HAE were amplified and sequenced using the primers listed in Figure 1 and primers spanning the NSVP genes between the termini ., The result showed an identical sequence with that of the HBoV1 Salvador isolate ( Genbank JQ923422 ) ., Additionally , no virus was detected in mock-infected B-HAE ( data not shown ) ., Taken together , these results demonstrate that the HBoV1 virions produced by pIHBoV1 transfection is capable of infecting polarized primary HAE cultures from cells derived from various donors and releasing identical progeny virions from infected primary HAE ., More importantly , we found that productive HBoV1 infection was persistent ., Although no gross cytopathic effects were observed in HBoV1-infected B-HAE , histology analysis of mock- vs . HBoV1-infected epithelia ( B33-11 ) revealed morphological differences: infected B-HAE did not feature obvious cilia at 7 days p . i . , and was significantly thinner than the mock-infected one on average at 22 days p . i . ( Figure 4B ) ., We further monitored the transepithelial electrical resistance ( TEER ) during infection of B-HAE , and found that at 6 days p . i . , it was reduced from a value of ∼1 , 200 to <400 Ω ., cm2 , while the mock-infected B-HAE maintained the initial TEER ( Figure 4C ) ., Notably , the decrease in TEER in the infected B-HAE was accompanied by an increase in HBoV1 secretion ( Figure 4A ) ., To confirm a role for HBoV1 infection in disruption of the barrier function of the epithelium , we examined the distribution of the tight junction protein Zona occludens-1 ( ZO-1 ) 39 ., Infected B-HAE showed dissociation of ZO-1 from the periphery of cells started from 7 days p . i . , compared with mock-infected B-HAE ( Figure 5A ) , which likely plays a role in reducing TEER ., Cumulatively , these results demonstrate that HBoV1 infection disrupts the integrity of HAE and that this may involve breakdown of polarity and redistribution of the tight junction protein ZO-1 ., To confirm a role for HBoV1 infection in the loss of cilia , we examined expression of the β-tubulin IV , which is a marker of cilia 40 , 41 ., In HBoV1-infected B-HAE , expression of β-tubulin IV was drastically decreased at 7 days p . i . , and was not detected at 22 days p . i . , in contrast to that in mock-infected B-HAE ( Figure 5B ) ., These results confirmed that HBoV1 infection caused the loss of cilia in infected B-HAE ., Notably , infected B-HAE showed changes of nuclear enlargement , which became obvious at 22 days p . i . ( Figure 5 , DAPI ) , indicating airway epithelial cell hypertrophy ., Collectively , we found that productive HBoV1 infection disrupted the tight junction barrier , lead to the loss of cilia and airway epithelial cell hypertrophy ., These are hallmarks of respiratory tract injury when a loss of epithelial cell polarity occurs ., Although primary HAE cultures support HBoV1 infection , their usefulness is limited by the variability between donors , tissue availability and high cost ., We thus explored alternative cell culture models for their abilities to support HBoV1 infection ., Using the purified HBoV1 , we examined HEK293 cells , other common epithelial cell lines permissive to common respiratory viruses 42 , including HeLa , MDCK , MRC-5 , LLC-MK2 and Vero-E6 , and several transformed or immortalized human airway epithelial cell lines ( A549 , BEAS-2B , 16HBE14o- 43 , NuLi-1 and CuFi-8 44 ) , as well as primary NHBE cells for the ability to support infection in conventional monolayer culture ., All were negative for HBoV1 infection as determined by IF analysis ( data not shown ) ., We next speculated that since some respiratory viruses infect polarized HAE but not undifferentiated cells 45 , some characteristics of the polarized epithelia may be critical for HBoV1 infection ., We thus polarized immortalized cells ( NuLi-1 and CuFi-8 ) at an air-liquid interface ( ALI ) for one month ., Once polarization was confirmed by detection of a TEER of >500 Ω ., cm2 , the cultures were infected with HBoV1 , under the same conditions as used for primary B-HAE cultures ., Notably , IF analysis revealed that at 10 days p . i . , HBoV1-infected CuFi-HAE ( differentiated from CuFi-8 cells ) was uniformly positive for NS1 ( Figure 6A ) , whereas the HBoV1-infected NuLi-HAE ( differentiated from NuLi-1 cells ) was not ( Figure S4 ) ., Moreover , the CuFi-HAE did express HBoV1 NS1 , NP1 and VP1/VP2 proteins ( Figures 6B and 6C ) ., The kinetics of virus release from the apical surface was similar to that of a primary B-HAE infected with virus at a similar titer ( maximally 2×107 gc/µl ) , although virus release from the basolateral surface was undetectable ( Figure 6D ) ., HBoV1 infection also resulted in a decrease in the thickness of the epithelium ( Figure 6E ) , and dissociation of the tight junction protein ZO-1 from the epithelial cell peripheries ( Figure 6F ) ., Collectively , these findings demonstrate that the immortalized cell line CuFi-8 44 , when cultured and polarized at an ALI , supports HBoV1 infection , and recapitulates the infection phenotypes observed in primary HAE , including destruction of the airway epithelial structure ., In this study , we have cloned the full-length HBoV1 genome and identified its terminal hairpins ., Virions produced from transfection of this clone into HEK293 cells are capable of infecting polarized HAE cultures ., Thus , we have established a reverse genetics system that overcomes the critical barriers to studying the molecular biology and pathogenesis of HBoV1 , using an in vitro culture model system of HAE ., It is notable that the HBoV1 terminal hairpins appear to be hybrid relicts of the prototype bocavirus BPV1 at the LEH , but of MVC at the REH 28 ., Replication of HBoV1 DNA in HEK293 cells revealed typical replicative intermediates of parvoviral DNA ., Although the head-tail junctions are unexpected in the replication of autonomous parvoviruses , they were likely generated during the cycle of rolling hairpin-dependent DNA replication 46 ., Therefore , we believe that the replication of HBoV1 DNA basically follows the model of rolling hairpin-dependent DNA replication of autonomous parvoviruses , with terminal and junction resolutions at the REH and LEH , respectively 46 ., The replication of parvoviral DNA depends on entry into S phase of the cell cycle or the presence of helper viruses 46 , 47 ., In this regard , it is puzzling that mature , uninjured airway epithelia are mitotically quiescent ( <1% of cells dividing ) 48–50 , as are the majority of the cells in polarized HAE ( in the G0 phase of the cell cycle ) ., However , recombinant adeno-associated virus ( AAV; in genus Dependovirus of the family of Parvoviridae ) infects HAE apically and expresses reporter genes 51–53 ., Gene expression by recombinant AAV requires a conversion of the ssDNA viral genome to a double-stranded DNA form that is capable to be transcribed 54 ., This conversion involves DNA synthesis ., Hence , we hypothesize that HBoV1 employs a similar approach to synthesize its replicative form DNA ., Notably , wild type AAV infected primary HAE apically and replicated when adenovirus was co-infected 55 ., The exact mechanism of how HBoV1 replicates in normal HAE will be an interesting topic for further investigation ., The airway epithelium , a ciliated pseudo-stratified columnar epithelium , represents the first barrier against inhaled microbes and actively prevents the entry of respiratory pathogens ., It consists of ciliated cells , basal cells and secretory goblet cells that together with the mucosal immune system , provide local defense mechanisms for the mucociliary clearance of inhaled microorganisms 56 ., The polarized ciliated primary HAE , which is generated by growing isolated tracheobronchial epithelial cells at an ALI for on average one month , forms a pseudo-stratified , mucociliary epithelium and displays morphologic and phenotypic characteristics resembling those of the in vivo human cartilaginous airway epithelium of the lung 57 , 58 ., Recent studies have revealed that this model system recapitulates important characteristics of interactions between respiratory viruses and their host cells 41 , 45 , 59–62 ., In the current study , we have examined primary B-HAE cultures obtained from three different donors ., HBoV1 infection of primary B-HAE was persistent and caused morphological changes of the epithelia , i . e . disruption of the tight barrier junctions , loss of cilia and epithelial cell hypertrophy ., The loss of the former , plasma membrane structures that seal the perimeters of the polarized epithelial cells of the monolayer , is known to damage the cell barrier necessary to maintain vectorial secretion , absorption and transport ., ZO-1 , which we monitored here , is specifically associated with the tight junctions and remains the standard marker for these structures ., Similarly , cilia play important roles in airway epithelia , in that they drive inhaled particles that adhere to mucus secreted by goblet cells outward 63 ., HBoV1 infection compromises barrier function , and thus potentially increases permeability of the airway epithelia to allergens and susceptibility to secondary infections by microbes ., The observed shedding of virus from the basolateral surface of infected primary HAE , albeit at a lower level ( ∼1 log lower than that from the apical surface ) , is consistent with the facts that HBoV1 infection disrupted the polarity of the pseudo-stratified epithelial barrier and resulted in the leakage to the basolateral chamber ., This explanation is also supported by HBoV1 infection of CuFi-HAE , where disruption of the tight junction structure was less severe and virus was released only from the apical membrane ., The induction of leakage by HBoV1 also suggests a mechanism that accounts for the viraemia observed in HBoV1-infected patients 5 ., Further disease pathology could be accounted for by infection-induced loss of cilia of the airway epithelia; a lack of cilia is often responsible for bronchiolitis 64–66 ., Therefore , our study provides direct evidence that HBoV1 is pathogenic to polarized HAE , which serves as in vitro model of the lung 57 , 58 ., Since HBoV1 is frequently detected with other respiratory viruses in infants hospitalized for acute wheezing 2 , 10–12 , the apparent pathological changes observed in HBoV1-infected HAE suggest that prior-infection of HBoV1 likely facilitates the progression of co-infection-driven pathogenesis in the patient ., The kinetics of virus release from the apical chamber of HAE infected with the progeny virus of pIHBoV1 ( cloned from the clinical Salvador1 isolate ) was similar to that following infection with the HBoV1 Bonn1 isolate , a clinical specimen 31 ., We believe that our study of HBoV1 infection of primary HAE reproduces infection of the virus from clinical specimens ., In addition , we generated virus from a pIHBoV1-b clone , which contains the NSVP genes from the prototype HBoV1 st2 isolate 1 ., Infection of primary B-HAE with this st2 virus resulted in a level of virus production similar to that observed here using the Salvador1 isolate ( data not shown ) ., We believe that our study with the laboratory-produced HBoV1 Salvador1 represents infection of HBoV1 of clinical specimens in HAE ., The MOI used for infection in the current study was high ., However , it should be noted that this titer is based on the physical numbers of virion particles as there are no practical methods for determining the infectious titer of HBoV1 preparations ., It should also be taken into consideration that extensive parvovirus inactivation occurs during the purification process , i . e . during CsCl equilibrium ultracentrifugation 67 ., Virus infection of HAE most likely reflects HBoV1 infection of the lung airways in patients with a high virus load in respiratory secretions 5 ., The fact that pIHBoV1 did not replicate well in undifferentiated human airway epithelial cells ( Figures 2B and 2C ) indicates that polarization and differentiation of the HAE is critical for HBoV1 DNA replication ., Nevertheless , polarized NuLi-HAE , which is derived from normal human airway epithelial cells , did not support HBoV1 infection , but the CuFi-HAE derived from airway epithelial cells isolated from a cystic fibrosis patient did ., The CuFi-HAE is unique relative to the others in that it retains the capacity to develop epithelia that actively transport in Na+ but not Cl− because of the mutation in the cystic fibrosis gene 44 ., Given the high complexity of the airway epithelium , we speculate that the permissiveness of HBoV1 infection is dependent on various steps of virus infection , e . g . attachment , entry , intracellular trafficking , and DNA replication of the virus ., Nevertheless , a polarized CuFi-HAE model derived from the CuFi-8 cell line represents a novel stable cell culture model that is providing unexpected insights into the infection characteristics of HBoV1 ., Although HBoV1 infection of CuFi-HAE reproduced disruption of the barrier tight junctions like that seen also in primary B-HAE , the absence of virus on the basolateral side implies that in HAE the secretion of HBoV1 is apically polarized ., We speculate that the milder damage of tight junctions in these cells might prevent virus release from the basolateral side of infected CuFi-HAE ., Further studies will focus on understanding the permissiveness of CuFi-HAE to HBoV1 infection and on the reason for the ease of infection of an HAE with a cystic fibrosis phenotype ., It has been shown that HBoV1 remains detectable in the upper airways of patients for weeks and months , even up to half a year 68–71 ., However , the mechanism behind this persistence , i . e . whether it is due to persistent replication and shedding , passive persistence after primary infection , or recurrent mucosal surface contamination , has remained unknown ., Our results in in vitro HAE cultures showed that HBoV1 is able to replicate and shed from both the apical and basolateral surfaces at least for three weeks , supporting the notion that shedding of the virus from the airways is a long-lasting process ., This may further explain why a high rate of co-infection , or co-detection , between HBoV1 and other respiratory viruses has been reported 5 ., Since recombinant AAV persists as an episome in transduced tissues , which prolongs gene expression 72 , 73 , it is possible that also the HBoV1 genome can be presented as an episome 29 , 30 for long term expression and replication ., Apparently , the mechanism underlying this feature of HBoV1 infection warrants further investigation ., However , in contrast to the other human-pathogenic B19V , HBoV1 does not seem to persist in human tissues for many years 74 ., In conclusion , our findings indicate that the innovative reverse genetics system for studying HBoV1 infection that we describe here will enable us to elucidate the mechanism of HBoV1 replication and pathogenesis in a polarized HAE ., Our system mimics natural HBoV1 infection of the in vivo human cartilaginous airway epithelia ., The pathogenesis of HBoV1 in co-infection with other respiratory viruses and in conditions of lung diseases is a focus of future study ., A nasopharyngeal aspirate was obtained from a child with community-acquired pneumonia in Salvador , Brazil , who had an acute HBoV1 infection ( seroconversion , viraemia , and over 104 gc of HBoV1 per ml of aspirate ) 17 ., Viral DNA was extracted according to a method described previously 77 ., The sequence of the head-to-tail junction of the HBoV1 episome suggests that HBoV LEH and REH share similarities both in structure and sequence with that of the BPV LEH and MVC REH , respectively 27 , 29 ., Based on this information 28 , we designed primers to amplify the HBoV1 termini , which are shown in Table 1 and Figure 1 ., The Phusion high fidelity PCR kit ( NEB , Ipswich , MA ) was used following the manufactures instructions , to amplify the left-end hairpin ( LEH ) and the right-end hairpin ( REH ) of HBoV1 ., Briefly , the DNA denaturation at 98°C for 30 s was followed by 35 cycles of: denaturing at 98°C for 10 s; annealing at 55°C for 15 s; and extension at 72°C for 30 s ., Following the final cycle , extension was continued at 72°C for 10 min ., The PCR products were analyzed by electrophoresis in a 2% agarose gel ., DNA bands were extracted using the QIAquick gel extraction kit ( Qiagen , Valencia , CA ) , and the extracted DNA was directly sequenced at MCLAB ( South San Francisco , CA ) , using primers complementary to the extended sequences on the forward and reverse amplification primers ., PCR-generated DNA was cloned in pGEM-T vector ( Promega , Madison , WI ) , and DNAs isolated from cultures of individual clones were subsequently sequenced ., Cells grown in 60-mm dishes were transfected with 2 µg of plasmid as indicated in Figure 2; the Lipofectamine and Plus reagents ( Invitrogen/Life Technologies , Carlsbad , CA ) were used as previously described 79 ., For some of the transfection experiments , HEK293 cells were cotransfected with 2 µg of pHelper plasmid ( Agilent ) , which contains the adenovirus 5 ( Ad5 ) E2a , E4orf6 , and VA genes , or infected with adenovirus type 5 ( Ad ) at an MOI of 5 as previously described 79 ., Low molecular weight ( Hirt ) DNA was extracted from transfected cells , digested with DpnI ( or left undigested ) and analyzed by Southern blotting as previously described 80 ., Cells were lysed , separated by SDS-8% polyacrylamide gel electrophoresis ( PAGE ) , and blotted with antibodies as indicated as previously described 81 ., HEK293 cells were cultured on fifteen 150-mm plates in DMEM-10%FCS , and transfected with 15 µg of pIHBoV1 per dish using LipoD293 ( SignaGen , Gaithersburg , MD ) ., After being maintained for 48 h at 5% CO2 and 37°C , the cells were collected , resuspended in 10 ml of phosphate buffered saline , pH7 . 4 ( PBS ) , and lysed by subjecting them to four freezing ( −196°C ) and thawing ( 37°C ) cycles ., The cell lysate was then spun at 10 , 000 rpm for 30 min ., The supernatant was collected and assessed on a continuous CsCl gradient ., In brief , the density was adjusted to 1 . 40 g/ml by adding CsCl , and the sample was loaded into an 11-ml centrifuge tube and spun in a Sorvall TH641 rotor at 36 , 000 rpm , for 36 h at 20°C ., Fractions of 550 µl ( 20 fractions ) were collected with a Piston Gradient Fractionator ( BioComp , Fredericton , NB , Canada ) , and the density of each was determined by an Abbes Refractometer ., Viral DNA was extracted from each fraction and quantified with respect to the number of HBoV1 gc , using HBoV1-specific qPCR as described below ., Those fractions containing the highest numbers of HBoV1 gc were dialyzed against PBS , and then viewed by electron microscope and used to infect HAE cultures ., The final purified virus preparation was concentrated by ∼5-fold , and adsorbed for 1 min on a 300-mesh copper EM grid coated with a carbon film , followed by washing with deionized water for 5 s and staining with 1% uranyl acetate for 1 min ., The grid was air dried , and was inspected on a 200 kV Tecnai F20 G2 transmission electron microscope equipped with a field emission gun ., Fully differentiated primary B- ( each of the three distinct subtypes ) , CuFi- and NuLi-HAE were cultured in Millicell inserts ( 0 . 6 cm2; Millipore ) and inoculated with 150 µl of purified HBoV1 ( 1×107 gc/µl in phosphate buffered saline , pH7 . 4; PBS ) from the apical surface ( at a multiplicity of infection , MOI , of ∼750 gc/cell; an average of 2×106 cells per insert ) ., For each of the HAE , a 2-h incubation was followed by aspiration of the virus from the apical chamber and by three washes of the cells with 200 µl of PBS to remove unbound virus ., The HAEs were then further cultured at an ALI ., For conventional monolayer cells , cells cultured in chamber slides ( Lab-Tek II; Nalge Nunc ) were infected with purified HBoV1 at an MOI of 1 , 000 gc/cell ., After HBoV1 infection , ALI membranes were fixed with 3 . 7% paraformaldehyde in PBS at room temperature for 15 min ., The fixed membranes were cut into several small pieces , washed in PBS three times for 5 min , and permeabilized with 0 . 2% Triton X-100 for 15 min at room temperature ., The membranes were then incubated with primary antibody at a dilution of 1∶100 in PBS with 2% FCS for 1 h at 37°C ., This was followed by incubation with a fluorescein isothiocyanate- or rhodamine-conjugated secondary antibody ., Confocal images were taken with an Eclipse C1 Plus confocal microscope ( Nikon , Melville , NY ) controlled by Nikon EZ-C1 software ., Primary antibodies used were anti- ( HBoV1 ) NS1

Introduction, Results, Discussion, Materials and Methods

Human bocavirus 1 ( HBoV1 ) has been identified as one of the etiological agents of wheezing in young children with acute respiratory-tract infections ., In this study , we have obtained the sequence of a full-length HBoV1 genome ( including both termini ) using viral DNA extracted from a nasopharyngeal aspirate of an infected patient , cloned the full-length HBoV1 genome , and demonstrated DNA replication , encapsidation of the ssDNA genome , and release of the HBoV1 virions from human embryonic kidney 293 cells ., The HBoV1 virions generated from this cell line-based production system exhibits a typical icosahedral structure of approximately 26 nm in diameter , and is capable of productively infecting polarized primary human airway epithelia ( HAE ) from the apical surface ., Infected HAE showed hallmarks of lung airway-tract injury , including disruption of the tight junction barrier , loss of cilia and epithelial cell hypertrophy ., Notably , polarized HAE cultured from an immortalized airway epithelial cell line , CuFi-8 ( originally derived from a cystic fibrosis patient ) , also supported productive infection of HBoV1 ., Thus , we have established a reverse genetics system and generated the first cell line-based culture system for the study of HBoV1 infection , which will significantly advance the study of HBoV1 replication and pathogenesis .

Human bocavirus 1 ( HBoV1 ) has been identified as one of the etiological agents of wheezing in young children with acute respiratory-tract infections ., HBoV1 productively infects polarized primary human airway epithelia ., However , no cell lines permissive to HBoV1 infection have yet been established ., More importantly , the sequences at both ends of the HBoV1 genome have remained unknown ., We have resolved both of these issues in this study ., We have sequenced a full-length HBoV1 genome and cloned it into a plasmid ., We further demonstrated that this HBoV1 plasmid replicated and produced viruses in human embryonic kidney 293 cells ., Infection of these HBoV1 progeny virions produced obvious cytopathogenic effects in polarized human airway epithelia , which were represented by disruption of the epithelial barrier ., Moreover , we identified an airway epithelial cell line supporting HBoV1 infection , when it was polarized ., This is the first study to obtain the full-length HBoV1 genome , to demonstrate pathogenesis of HBoV1 infection in human airway epithelia , and to identify the first cell line to support productive HBoV1 infection .

virology, viral nucleic acid, emerging viral diseases, biology, microbiology, viral replication

journal.pgen.1000568

Evolution and Survival on Eutherian Sex Chromosomes

Therian sex chromosomes , X and Y , evolved from a pair of homologous autosomes and thus originally harbored an identical set of genes 1–3 ., Driven by a male-determining locus ( SRY ) , the stepwise suppression of recombination between the Y and the X led to evolutionary strata corresponding to individual suppression events 1 ., Suppression of recombination between the Y and the X also resulted in their current dramatically different gene numbers 2 , ∼1 , 100 and <200 genes on the human X and Y , respectively 4 , 5 ., While many X-linked genes have been preserved , the majority of Y-linked genes have been pseudogenized or deleted ., Purifying selection is predicted to be inefficient in nonrecombining regions of the Y , causing an accumulation of deleterious mutations; eventually , genes are expected to be lost by means of Mullers ratchet , background selection , the Hill-Robertson effect , and/or genetic hitchhiking of beneficial mutations 6 , 7 ., The already gene-poor mammalian Y continues to deteriorate 8 , and it has been proposed that within a few million years the human Y will lose all of its genes , with major consequences for mankind 2 , 9 ., The human Y has retained a meager 16 functional single-copy protein-coding genes described as X-degenerate 10 , i . e . possessing divergent X chromosome gametologs ( gametologs are X-Y homologs 11 ) ., Therefore , these genes represent relics of ancient autosomal genes ( the remaining functional Y-linked genes are classified as pseudoautosomal , ampliconic , and recently X-transposed 5 ) ., What evolutionary forces have been maintaining these X-degenerate genes on the Y ?, The first possibility is that the surviving genes might carry out essential functions where purifying selection maintains the amino acid sequence of the encoded protein leading to a low rate ratio of nonsynonymous to synonymous substitutions ( KA/KS ) ., However , decreased efficacy of such selection on the Y would elevate KA/KS for Y vs . X gametologs 8 ., The second possibility is that recombination suppression between the X and the Y can be viewed , effectively , as a duplication event ., There are several proposed scenarios for how paralogs diverge from one another , including asymmetric evolution , where one copy is presumed to maintain the ancestral function , and thus experiences stronger purifying selection , while the other copy can undergo neofunctionalization or pseudogenization 12 and thus might experience positive selection or evolve neutrally ., If this scenario holds true with respect to X and Y divergence , we expect that X gametologs will maintain the ancestral somatic functions necessary to both males and females ( because the X is present in both sexes ) , and will evolve under purifying selection ., Purifying selection might be strong on the X because it is hemizygous in males and thus recessive alleles are readily available for such selection to operate there ., Y-linked genes , present only in males may undergo neofunctionalization , or , as has often been observed , may undergo pseudogenization 4 , 5 , 10 ., Purifying selection is expected to be weak for genes on the Y because of the lack of recombination there ( see above ) ., Thus , similar to paralogs , divergence in function and expression between Y- and X-gametologs might actually contribute to the survival , in addition to the accelerated evolution 13 , of Y chromosome genes ., Previous studies have observed elevated evolutionary rates for Y- versus X-linked genes ., For instance , evolutionary rates were found to be higher for human and mouse Y chromosome genes compared with their gametologs on the X 13 ., However , without available outgroup sequences , the incipient stages of X- and Y-linked gene evolution remained ambiguous , i . e . , the ancestral sex chromosome branch could not be broken into X- and Y-specific segments ., In a different study , not only was purifying selection shown to be less potent in exons of three primate Y than X chromosome genes , but positive selection was also evident at several sites of Y chromosome exons 8 ., Nevertheless , as both sex chromosomes carry genes with a nonrandom assortment of functions ( e . g . , genes involved in spermatogenesis are enriched on the Y 14 , whereas genes important for reproduction and brain function are overrepresented on the X 2 ) , contrasting only the X- and Y-linked genes might not represent an ideal way to study the evolution of either gene group ., When feasible , a direct comparison of sex chromosome genes with homologous autosomal genes is therefore warranted ., Tied to the understanding of sex chromosome evolution are hypotheses of how X and Y diverged from each other forming different evolutionary strata ., Each stratum corresponds to a distinct recombination suppression event , thus , gametologs belonging to the same stratum have similar divergence 1 ., In eutherian mammals , five strata of increasing age are observed linearly along the X chromosome , with the youngest near its proximal end and the oldest near its distal end , suggesting that suppression of recombination occurred in a stepwise manner between X and Y 1 , 4 ., The arrangement of homologous sequences on the Y chromosome has been scrambled , supporting the hypothesis about the role of inversions in Y chromosome evolution 1 , 4 ., While some X-degenerate Y chromosome genes were retained from the original autosomal pair , others were added later ., After eutherian-marsupial divergence ( ∼166 MYA 15 ) , the eutherian sex chromosomes acquired the X-/Y-added region ( XAR/YAR ) , through a translocation from an autosome 16 ., This segment remains autosomal in marsupials and monotremes 16 , 17 and provides a direct comparison of homologous genes between autosomes and sex chromosomes ., Such a comparison allows us to infer the eutherian proto-sex chromosome branch and separate the ancestral sex chromosome branch into X- and Y-specific portions , i . e . to investigate emergent eutherian sex chromosome evolution ., In eutherian mammals , the XAR/YAR continued to recombine between X and Y until the formation of strata 3 and 4 , app roximately 80–130 MYA and 30–50 MYA , respectively 1 ., Primates and rodents diverged ∼85–90 MYA 18 , and thus genes belonging to stratum 3 putatively began evolving as X- and Y-specific in the ancestor of eutherian mammals ., It is expected that stratum 4 genes only evolved as X- and Y-specific along the primate lineage ., Only 12 human gametologous pairs with functional Y homologs are left in the human XAR/YAR 1 , 4: TMSB4X/Y , CX/YORF15A , CX/YORF15B , EIF1AX/Y , ZFX/Y , USP9X/Y , DDX3X/Y , and UTX/Y are classified in stratum 3 1 , 4; but there has been some debate whether stratum 4 contains PRKX/Y , NLGN4X/Y , TBL1X/Y , and AMELX/Y ( classified based on sequence divergence 1 ) or whether TBL1X/Y and AMELXY/Y belong , instead , to stratum 3 ( based on analysis of parsimonious inversions 4 ) ., Here , in our attempt to analyze the early stages of sex chromosome evolution , as well as to address what evolutionary forces lead to preservation of functional Y chromosomal gametologs , we analyzed 12 XAR/YAR gametologous pairs in eutherians along with their autosomal orthologs in opossum and platypus . A direct comparison of homologs decreased biases due to sequence composition , gene size , and ancestral functional constraints possible in studies juxtaposing Y- and X-linked genes against nonhomologous autosomal genes ( e . g . , 19 ) ., Specifically , we tested the following hypotheses:, 1 ) whether X and Y evolved unique evolutionary rates immediately after the suppression of recombination between them;, 2 ) whether the evolutionary rates along both the X and Y branches have been constant throughout their evolutionary histories , and ,, 3 ) whether gametolog evolution parallels paralog evolution in terms of rates and functional constraints ., Additionally , by utilizing whole-genome transcriptome and other published experimental data , we examined whether the expression and functional divergence of Y from X gametologs correlated with their evolution and potentially contributed to their survival on the sex chromosomes ., Because of the use of opossum and platypus sequences , for the first time we are able to get a glimpse of how the ancestral eutherian sex chromosomes evolved ., To test the hypotheses stated above , we studied the evolution of all 12 available XAR/YAR human functional gametologs 4: PRKX/Y , NLGN4X/Y , TBL1X/Y , AMELX/Y , TMSB4X/Y , CX/YORF15A , CX/YORF15B , EIF1AX/Y , ZFX/Y , USP9X/Y , DDX3X/Y , and UTX/Y , here listed starting from the Xpter ( Figure 1; the Y-linked gametolog of CXorf15 in human and chimpanzee has been split into two genes , CYorf15A and CYorf15B 10 , which we investigate separately ) ., We included sequences from eight eutherian mammals ( human , chimpanzee , rhesus , horse , cow , dog , mouse and rat ) that had sufficient sequence coverage for robust analysis of all of the genes in the XAR ( Figure 2 , Figure 3 , and Materials and Methods ) as well as human , chimpanzee and ( when available ) mouse YAR gene sequences ., To isolate chromosome-specific effects and to delineate the ancestral and proto-sex chromosomes branches , we included the orthologous autosomal gene sequences from opossum and platypus ., In opossum , the order of genes found in the XAR/YAR is the same as in eutherians , but the sequences are split between chromosomes 4 and 7 20 ., The platypus genome is not yet assembled , however , the presence of the orthologous genes on a single chicken chromosome ( chromosome, 1 ) 4 , in the same order , suggests that the original translocation likely occurred in one event ., The phylogenetic analysis of the coding region within each homologous XAR/YAR gene group usually resulted in one of two separate tree topologies ., For DDX3X/Y , USP9X/Y , and UTX/Y , we observed the pre-radiation tree topology ( Figure 1 , Figure 2 , Figure S1 ) , in which X- and Y-linked genes formed two distinct clades , and thus these gametologs diverged from one another in the common ancestor of boreoeutherian mammals 21 , forming stratum 3 , believed to be shared among all eutherian mammals 1 ., For PRKX/Y , NLGN4X/Y , TBL1X/Y , AMELX/Y , and TMSB4X/Y , we observed the post-radiation tree topology ( Figure 1 , Figure 3 , Figure S1 ) , in which primate gametologs clustered together , and therefore recombination suppression between them followed the boreoeutherian radiation and presumably occurred along the primate lineage , forming stratum 4 ., For genes with the post-radiation topology , consistent with previous experimental assays 22–24 , we did not identify the homologous mouse Y genes , suggesting that they have been deleted , pseudogenized beyond the recognition of the alignment algorithms utilized , or are yet unsequenced ( Materials and Methods ) ., For each gene with either the pre- or post-radiation topology , the observed topology was significantly different from the alternative topology ( Table S1 ) ., Genes for which the topology could not be confidently determined , CX/Yorf15A , CX/Yorf15B , EIF1AX/Y and ZFX/Y ( Figure S1 ) , were excluded from the concatenated analysis ( Table S1 ) , along with NLGN4X/Y ( Figure S1 ) , because its murid X orthologs could not be identified reliably 25 ., To test for gene conversion , we conducted a phylogenetic analysis of each exon individually ., Exons where the X and Y sequence from the same species formed a unique clade have putatively undergone gene conversion and were excluded from further analysis ( Table S2 ) ., In most cases though , the phylogenetic trees produced for each exon were identical to the topology of the parent gene ., When exons following the post- and pre-radiation topology were mapped to the X chromosome , they grouped closest and furthest from the Xpter , respectively ( Figure, 1 ) in a significantly non-random distribution ( P<2 . 2×10−16; Wilcoxon rank-sum test ) ., Although gene conversion was detected for isolated exons ( Table S2 ) , the observed distribution is more parsimoniously explained by two distinct evolutionary strata ., Thus , either the boundary separating strata 3 and 4 , is closer to the position suggested in 1 , i . e . between TMSB4X and AMELX , or it is located between TBL1X and NLGN4X , as proposed in 4 , but stratum 3 should be split into two sub-strata with a second boundary somewhere between USP9X and TMSB4X ( Figure 4 ) ., Homologous marsupial and monotreme sequences have allowed us to expand upon previous efforts investigating sex chromosome evolution 13 ., In particular , for the pre-radiation topology , we were able to separate the ancestral sex chromosome branch ( preceding the boreoeutherian divergence ) into X- and Y-specific portions ( labeled Ancestral X and Ancestral Y , respectively , Figure 2A ) and to delineate the eutherian proto-sex chromosome branch ( labeled Proto-Sex , Figure 2A ) , preceding the Y chromosome inversion that led to formation of stratum 3 ., Similarly , for primates in the post-radiation topology , we were able to investigate the evolution of X- and Y-linked sequences before ( identified by the Proto-SexPrimate branch ) and after the recombination suppression event that led to the formation of stratum 4 ( indicated on the AncestralPrimateX and AncestralPrimateY branches ) ., To study differences in evolutionary rates of X , Y , and autosomal genes , we concatenated the coding regions of genes following the pre-radiation ( PRKX/Y , TBL1X/Y , AMELX/Y and TMSB4X/Y; a total of 2700 bp ) and post-radiation ( USP9X/Y , DDX3X/Y and UTX/Y; a total of 6108 bp ) topology separately ( Materials and Methods , Table S1; bootstrap values shown in Figure S2 ) , to reduce the confounding influences of comparing genes from potentially different strata ., Further , we masked out exons from the exon-by-exon analysis described above that ( 1 ) did not conform to the topology characteristic for the majority of the exons of the gene ( these are likely gene conversion events ) , ( 2 ) produced an ambiguous tree topology , or ( 3 ) lacked sufficient data ( see Materials and Methods ) ., First , we investigated how synonymous rates differ among the two sex chromosomes and the homologous autosomal sequence ., Synonymous rates for genes with the pre-radiation topology ( Figure, 2 ) were significantly higher for Y than X gametologs ( between the sum of branches to the common ancestor between human X and Y , P\u200a=\u200a1 . 01×10−3; chimpanzee X and Y , P\u200a=\u200a1 . 31×10−3; and mouse X and Y , P\u200a=\u200a4 . 40×10−6 ) , reflecting male mutation bias 26 ., Genes with this topology had significantly higher synonymous rates for mouse than human ( compared between the sum of branches to the common ancestor , P\u200a=\u200a2 . 43×10−10 for mouse X - human X , P\u200a=\u200a2 . 54×10−10 for mouse Y - human Y ) , in agreement with previous studies ( e . g . , 27 ) ., Synonymous rates for genes with the post-radiation topology ( Figure 2B ) were ( not significantly ) higher between mouse X vs . human X , and similar between human Y and X sums of branches ( data not shown ) ., Synonymous rates were lower in the opossum lineage ( 0 . 282 and 0 . 530 for the pre- and post-radiation topology , respectively ) than in even the shortest eutherian lineages ( 0 . 469 and 1 . 227; calculated as the sum of eutherian-specific branches leading to Human X for the pre-radiation topology and Horse X for the post-radiation topology , respectively ) ., This can be explained by the lower GC content and reduced recombination rates of opossum vs . eutherian chromosomes 20 , 28 ., The differences in opossum rates between the pre- and post-radiation topologies might either result from interchromosomal rate variation 29 , since most of the genes following the former and latter topologies are found on opossum chromosomes 4 and 7 , respectively , or be driven by local genomic factors 30 ., Second , we studied variation in the KA/KS ratio among branches ., For every comparison in both topologies , the KA/KS ratio was significantly higher for the Y than the X branch ( Figure 2B , Figure 3B ) ., Our data set allowed us to investigate when these differences between X and Y chromosome evolution emerged , i . e . whether the elevated evolutionary rates observed on the Y versus the X occurred immediately after recombination suppression or just recently , after million years of suppressed recombination ., For both topologies , immediately after recombination suppression , the Y chromosome ( Ancestral Y and Ancestralprimate Y branches for pre- and post-radiation , respectively ) acquired significantly higher KA/KS ratios as compared with the Proto-Sex branch ( Figure 2B , Figure 3B ) ., This increase could be due to relaxed purifying selection on the Y in the absence of recombination and/or due to positive selection of Y-linked genes that acquired new functions 8 ., Positive selection was not detected on any branches or sites in these seven genes ( see Materials and Methods ) and , consequently , KA/KS ratios were interpreted as varying degrees of purifying selection , reflecting the level of functional constraints ., Thus , purifying selection was weaker on the Ancestral Y branch than on the Proto-Sex branch ( or the Ancestral X branch ) for trees with both topologies ( Figure 2B , Figure 3B ) ., In contrast , the intensity of purifying selection did not differ significantly between the Proto-Sex and Ancestral X branches for gametologs following the pre-radiation topology , implying that these X-linked genes have retained the level of functional constraints of their autosomal ancestors ( Figure 2B ) ., Interestingly , X and Y lineages of the pre-radiation topology maintained relatively constant KA/KS ratios since the suppression of recombination between them ( Figure 2B; recent gametolog separation in the post-radiation topology prevented us from conducting a similar analysis there ) ., Indeed , the KA/KS ratio was not significantly different between the Ancestral X branch and either the ape or the mouse X branches , again suggesting preservation of functional constraints of X gametologs ., Similarly , the KA/KS ratio did not differ significantly between the Ancestral Y branch and either the ape or the mouse Y branches , indicating that Y rapidly settled on its own equilibrium evolutionary rate 13 ., We next asked whether divergence between gametologs mimicked the divergence between paralogs ., To answer this question , we compared the evolution of human gametologs ( here all 12 gametologous pairs were considered ) against pairs of similarly aged human autosomal paralogs ., Using the synonymous rate ( KS ) as an estimate of evolutionary age , for each gametolog , we compiled a set of similarly aged autosomal trios composed of a pair of human paralogs , duplicated after human-opossum divergence , aligned with the orthologous autosomal sequence in opossum ( a total of 470 trios; Materials and Methods ) ., The distribution of pairwise KA/KS ratios was significantly different between gametologs and similarly aged autosomal paralogs ( P\u200a=\u200a0 . 0001 , Wilcoxon test ) ., The impact of positive selection was minor ( only 13 sites of CYorf15B and 5 sites of ZFY exhibited signatures of positive selection; Materials and Methods ) , and thus we again interpreted the KA/KS ratio as the strength of purifying selection ., Pairwise KA/KS ratios were lower for nine out of 12 gametologs than for autosomal paralogs ( Table 1 ) , suggesting stronger purifying selection acting on gametologs ., The higher pairwise KA/KS ratios observed for AMELX/Y , CX/Yorf15A and CX/Yorf15B might reflect the initial stages of Y gametolog pseudogenization 10 , 31 or positive selection acting on some CYorf15B sites ., Stronger purifying selection between most gametologs than paralogs contradicts the hypothesis of sexual selection driving more rapid divergence between gametologs than autosomal paralogs 32 ., Using opossum sequence to polarize substitutions , we determined that most gametologs displayed asymmetric functional constraints , meaning that the KA/KS ratios differed between the two gametologs , often by an order of magnitude , although not always significantly so , and all gametologs had a lower KA/KS ratio for the X than Y copy ( Table 1 ) ., Thus , gametologs likely followed an evolutionary scenario proposed for paralogs , in which purifying selection was stronger for one than the other paralogous copy 12 ., And , consistent with our expectation ( see Introduction ) , purifying selection was always stronger for the X than the Y copy ., We next asked whether X and Y gametologs evolved at rates similar to these for slowly and quickly evolving paralogous copies , respectively ( slowly and quickly evolving paralogous copies were determined using opossum as an outgroup ) ., In contrast to expectations of inefficient purifying selection on the Y 6 , all but three Y gametologs had lower KA/KS ratios and thus may have evolved under stronger purifying selection than the quickly evolving copies of paralogs ( Table 1 ) ., This might represent a mechanism of Y gametolog preservation; either a gene must be maintained through purifying selection , or , as evident again for AMELY , CYorf15A , and CYorf15B , that had higher KA/KS ratios than the similarly aged quickly evolving paralogs , they may become prey to pseudogenization ., Relatively strong purifying selection observed for Y gametologs might also , in part , be explained by genetic hitchhiking due to selection acting on other Y chromosome genes ( e . g . , ampliconic genes ) ; genetic hitchhiking is expected to be particularly potent on the Y because it does not undergo recombination outside of the pseudoautosomal regions ., Similar to Y gametologs , all but two X gametologs had lower KA/KS ratios than the slowly evolving paralogous copies ( Table 1 ) ., Intense purifying selection acting on X gametologs can be explained by the fact that X is hemizygous in males ( thus recessive alleles are instantly open to selection ) and by the preservation of somatic functions important for both sexes ., To test a hypothesis that the expression and functional divergence of Y gametologs from their X counterparts potentially contributed to the survival of the former on the sex chromosomes , we compiled and analyzed whole-genome transcriptome and other published experimental data ., Expression divergence between X and Y gametologs was inferred from human and mouse transcriptome microarray data produced by Su and colleagues 33 ., In humans we studied 11 tissue samples collected from males in that study ., In over three quarters of gametolog-tissue combinations , either the X and Y gametologs in a pair were expressed at unequal levels ( at least 25% different ) or one copy was completely turned off ( Figure 5 ) ., Thus , gametologs acquired expression patterns distinct from one another ., We found no significant difference in the expression divergence between human gametologous pairs and similarly aged human autosomal paralogs ( Table S3 ) , implying that the expression patterns of gametologous pairs diverge from one another at a similar rate as those of paralogous pairs ., Next , using the proportion of tissues in which both the X and Y gametolog are similarly expressed ( white boxes with a number in Figure 5 ) among all tissues with detected expression as a measure of gametolog expression similarity , we determined that there is no significant difference in expression patterns between gametologs following the pre- versus post-radiation topologies ( Wilcoxon rank sum test , P\u200a=\u200a0 . 3018 ) , and there is no significant correlation ( P\u200a=\u200a0 . 622 ) between gametolog expression similarity and the distance from the Xpter ., The non-significance may be due to both the limited number of genes , as well as the limited number of tissues available for the analysis ., However , given that expression patterns diverge very rapidly , frequently outpacing sequence divergence 34 , 35 , the genes considered here may already have diverged past any threshold of observing certain correlations ., Mouse samples used in the study of Su and colleagues 33 , were all pooled from tissues collected from both males and females , thus it was impossible to distinguish levels of X and Y expression unambiguously ., Still , two mouse Y-linked genes included in microarrays analyzed by Su and colleagues 33 - Ddx3y and Usp9y - had undetectable expression across all 61 tissues analyzed , while their gametologs , Ddx3x and Usp9x were expressed in all and one of the tissues examined , respectively ( the other gametologs present on the array studied , Utx/y and Zfx/y , were not expressed 33 ) ., Therefore , we do observe unique expression patterns between at least some mouse and most human X and Y gametologs ., These differences in expression might have contributed to the retention of Y gametologs ., Additionally , mining and compiling nearly 15 years of experimental data gathered from the literature allowed us to conclude that the majority of human X and Y gametologs acquired unique protein expression patterns and/or functions ( Table S4 ) , sometimes not detectable from studies of gene expression alone ., For instance , in the case of human DDX3X/Y , although both gametologs are widely transcribed , only the X-linked copy , DDX3X , is also widely translated , while DDX3Y is translated exclusively in the male germ line 36 ., This is accompanied by distinct temporal protein expression patterns , at least in spermatogenesis , where the two protein products are present at different stages 36 ., In another example , the TBL1X/Y gametologs differ in both mRNA expression and protein function ., TBL1X mRNA is ubiquitously expressed 37 , while TBL1Y mRNA expression is limited to only a few tissues 38 ., The dissimilarity is also evident in function as the TBL1X protein represses transcription 39 , while the TBL1Y protein has no such activity 38 ., As a final example , AMELY deletions cause no detectable phenotypic changes 40 , but deletion of AMELX causes amelogenesis imperfecta 31 , 41 ., Such differences in protein expression and function between gametologs might have also contributed to retention of X degenerate Y chromosome genes ., To the best of our knowledge , we present the first analysis of the ancestral proto-sex evolutionary rates in eutherian mammals ., We observed that immediately following the suppression of recombination between X and Y , likely due to their importance in both sexes , X gametologs largely maintained the ancestral autosomal sequence and functional constraints ., In contrast , Y gametologs , as predicted due to absence of recombination 6 , evolved under weaker purifying selection than X gametologs ., Further , these different rates have been roughly maintained through evolutionary time by each of the sex chromosomes ., Both X and Y gametologs , on average , acquired functional constraints stronger than quickly and slowly evolving copies of autosomal paralogs , respectively ., This might have contributed to the survival of these gametologs ., We also observe that the divergence between of X and Y gametolog sequences after recombination suppression , in some ways mimics that of paralogous genes , were one copy maintains a lower , more conservative , rate of evolution while the other is allowed a higher substitution rate , and may eventually evolve a new function or become prey to pseudogenization ., Our analysis of the sequence evolution combined with experimental observations suggests that to withstand the evolutionary vulnerability on the Y chromosome , most Y-linked genes diverged in expression and function from their X gametologs to become separately valuable ., Although Y chromosome sequencing and assembly is an undeniably challenging endeavor 5 , 10 , 42 , it provides invaluable and otherwise impossible insights into mammalian evolution ., Further studies investigating gametologs will critically depend on the availability of Y chromosome sequences for several mammals , in addition to human 5 and chimpanzee 42 ., Eutherian X-linked and corresponding autosomal nucleotide sequences for opossum and platypus were extracted from the 28-way vertebrate alignments 43 available through Galaxy 44 , using the human X homolog as a reference ., We initially considered X-linked sequences from all 18 eutherian species included in the 28-way genomic alignments 43 , but retained only eight due to limited coverage in the other species ( Figure 2 and Figure 3 ) ., Only complete human and chimpanzee Y 5 , 10 , and partial mouse Y chromosome sequences are available ., Human , chimpanzee and mouse Y-linked sequences were downloaded from Genbank ( see Table S5 ) ., Of the 12 gametologs , we identified only four ( Zfy , Usp9y , Ddx3y , and Uty ) annotated on the mouse Y chromosome in Genbank ., Since the mouse Y chromosome has yet to be completely sequenced and assembled , we searched the available 533 mouse Y BACs ( a total of ∼90 Mb ) for the seven missing genes ., Using BlastZ 45 , we identified the four previously annotated genes ( see above ) , but were unable to locate the unannotated genes ., The coding nucleotide sequences for each homologous gene group ( sex-linked gametologs and autosomal homologs ) were aligned using ClustalW 46 ., The phylogenetic trees were built according to the Neighbor-Joining method 47 as implemented in PHYLIP 48 using X-linked sequences from human , chimpanzee , rhesus , mouse , rat , cow , dog , horse , Y-linked sequences from human , chimpanzee , and mouse , when available , and autosomal sequences from opossum and platypus ., These species were chosen among the 18 mammals represented in 43 because for each of them at least nine of the 12 genes had greater than 50% sequence coverage ., 1000 bootstrap replicates were generated first for each gene and then for each coding exon ., Exons with less than 50% bootstrap support for clades with either the pre- or post-radiation topology , fewer than 24 nucleotides aligned across all species , or inconsistent with the topology of the whole gene ( a total of 92 exons ) were excluded from this portion of the analysis ., In addition to Neighbor-Joining analysis , we used Maximum Likelihood and Maximum Parsimony tree building methods 48 ., The three approaches led to similar results ( data not shown ) ., Our results represent gene trees , not necessarily species trees ( see discussion of primate , rodent , and carnivore groupings in 49 ) , and so we advise against using these groupings to support arguments for or against contentious species groupings ., The exon by exon analysis described above led us to identify known cases of gene conversion ( e . g . in ZFX/Y 50 ) ., To further test for gene conversion , we aligned human X with human Y , chimp X with chimp Y and mouse X with mouse Y sequences using PipMaker 51 , a software that utilizes a local alignment algorithm to output regions of similar sequence identity ., Higher identity of a particular stretch of an alignment in relation to the entire alignment can be suggestive of gene conversion 52 ., New instances of gene conversion were not detected either with this method nor with GENECONV 53 ., HyPhy was used to estimate the branch-specific KS and KA under the GY94_3×4 model and to test for statistical significance of differences in the synonymous rates among branches using a Likelihood Ratio Test ( LRT ) , testing the likelihood that two branches had the same vs . different KS values 54 ., Tests conducted with the MG94_3×4 and MG94xHKY_3×4 models yielded similar statistically significant results ., To compute the probability that the KA/KS ratio was significantly different between two branches , we used the GAbranch analysis 55 in the online version of HyPhy ( www . datamonkey . org ) , which computes the model-averaged probability that two branches have the same KA/KS ratio 56 ., To determine the significance of the difference between sums of branches , we re-

Introduction, Results/Discussion, Materials and Methods

Since the two eutherian sex chromosomes diverged from an ancestral autosomal pair , the X has remained relatively gene-rich , while the Y has lost most of its genes through the accumulation of deleterious mutations in nonrecombining regions ., Presently , it is unclear what is distinctive about genes that remain on the Y chromosome , when the sex chromosomes acquired their unique evolutionary rates , and whether X-Y gene divergence paralleled that of paralogs located on autosomes ., To tackle these questions , here we juxtaposed the evolution of X and Y homologous genes ( gametologs ) in eutherian mammals with their autosomal orthologs in marsupial and monotreme mammals ., We discovered that genes on the X and Y acquired distinct evolutionary rates immediately following the suppression of recombination between the two sex chromosomes ., The Y-linked genes evolved at higher rates , while the X-linked genes maintained the lower evolutionary rates of the ancestral autosomal genes ., These distinct rates have been maintained throughout the evolution of X and Y . Specifically , in humans , most X gametologs and , curiously , also most Y gametologs evolved under stronger purifying selection than similarly aged autosomal paralogs ., Finally , after evaluating the current experimental data from the literature , we concluded that unique mRNA/protein expression patterns and functions acquired by Y ( versus X ) gametologs likely contributed to their retention ., Our results also suggest that either the boundary between sex chromosome strata 3 and 4 should be shifted or that stratum 3 should be divided into two strata .

Using recently available marsupial and monotreme genomes , we investigated nascent sex chromosome evolution in mammals ., We show that , in eutherian mammals , X and Y genes acquired distinct evolutionary rates and functional constraints immediately after recombination suppression; X-linked genes maintained lower , ancestral ( autosomal ) , rates , whereas the evolutionary rates of Y-linked genes increased ., Most X and , unexpectedly , Y genes evolved under stronger purifying selection than similarly aged autosomal paralogs ., However , we also observed that the divergence of gametologs and paralogs shared similar features ., In addition , many Y-linked copies evolved unique functions and expression patterns compared to their counterparts on the X chromosome ., Therefore , our results suggest that to be retained on the Y chromosome , genes need to acquire separately valuable expression and/or functions to be safeguarded by purifying selection .

evolutionary biology/bioinformatics, genetics and genomics/bioinformatics, computational biology/genomics

journal.ppat.1005161

Intrinsic MyD88-Akt1-mTOR Signaling Coordinates Disparate Tc17 and Tc1 Responses during Vaccine Immunity against Fungal Pneumonia

The rising incidence rate of life threatening fungal infections in immune-deficient hosts requires preventive measure in at risk individuals ., CD4+ T cells are the primary effector cells that control fungal infections in healthy hosts , and their loss in lymphopenic patients necessitates targeting residual immune subsets to elicit antifungal immunity ., We previously showed in a mouse model of lethal fungal pneumonia that , even in the absence of CD4+ T cell help , vaccine-induced CD8+ T cells could differentiate and expand into cytokine producing cells , persist as long-lasting memory cells , and mediate sterilizing immunity 1 ., Antifungal CD8+ T cells that produce IL-17A are indispensable in this model ., In contrast , CD8+ T cells that produce type I cytokines ( IFNγ , TNFα or GM-CSF ) contribute to vaccine immunity , but are expendable 2 , 3 ., A deeper understanding of the elements required to elicit CD8+ T cell responses will be required to catalyze the development of rationally designed anti-fungal vaccines ., T cell respond to antigen in three distinct phases: in the expansion phase , upon recognition of cognate antigen , T cells undergo rapid proliferation and differentiation into effectors; in the contraction phase , ~90% of effectors T cells die by apoptosis; and in the memory phase , the remaining 10% of effector T cells differentiate into long-lasting memory cells ., Hence , in general , the magnitude of expansion and survival of effector cells will dictate protective immunity 4 ., The inflammatory milieu influences the quality and quantity of effector T cells ., For example , a lack of type I interferon signaling abrogates clonal expansion of CD8+ T cells due to reduced survival , whereas enhanced inflammation exaggerates terminal differentiation and apoptosis 5 , 6 ., Among other factors , cytokines regulate differentiation of T cells into distinct subsets that express prototypic transcription factors and signature cytokines ., For Th17 cell responses , different combinations of cytokines including IL-6 , TGFβ , IL-1 , IL-21 and IL-23 have been implicated in differentiation in vitro and in vivo 7 , 8 ., CD8+ T cell responses are typically associated with defense against intracellular pathogens and tumors by mechanisms that are largely dependent on IFNγ , granzyme , and perforin ., CD8+ T cells control fungal infections chiefly by secretion of proinflammatory cytokines such as IFN-γ , TNF-α , and GM-CSF that activate phagocytes to kill fungi 9 ., A distinct subset of IL-17A producing CD8+ T cells , Tc17 cells , also play a role in defense against infections and tumors ., Elimination of Tc17 cells is associated with progressive SIV/HIV infection 10–12 and Tc17 cells are protective against vaccinia and influenza virus infections 13 , 14 and tumors 15 , 16 ., Likewise , we have found that Tc17 cells are indispensable for vaccine-induced protection against fungal pneumonia 2 ., Differentiation of Tc17 cells requires TGFβ and IL-6 or IL-21 17; IL-23 signaling has been shown to promote pathogenic Tc17 cells 18 ., IRF4 facilitates Tc17 responses by transcriptionally activating RORγt and RORα and repressing EOMES and FOXP3 , while IRF3 inhibits Tc17 programming by altering RORγt promoter binding 19 , 20 ., The molecular switch that regulates initial programming of Tc1 and Tc17 responses under similar in vivo ‘inflammatory milieu’ is poorly understood ., MyD88 , a signaling adaptor for TLRs and IL-1R family members in myeloid cells , is critical for innate and adaptive immunity 21 ., MyD88 signaling activates macrophages and DCs , elicits production of proinflammatory cytokines and promotes antigen presentation to initiate adaptive immune responses during viral , bacterial and parasitic infections 22 ., Impaired MyD88 signaling increases susceptibility to fungal infections such as candidiasis , cryptococcosis , aspergillosis , paracoccidioidosis , pneumocystis and coccidioidomycosis 23–25 ., Conversely , bolstering MyD88 signaling in dendritic cells improves resistance to aspergillosis 26 , 27 ., Thus , MyD88 signaling in myeloid cells plays an integral role in immunity against fungal infections ., However , the T cell-intrinsic role of MyD88 in adaptive immune responses to fungal infections has not been defined ., In experimental Toxoplasma gondii infection , T cell expression of MyD88 is required for Th1 mediated resistance 28 ., This Th1 response is independent of IL-1R and IL-18R , implying a role for TLRs in orchestrating MyD88-mediated T cell responses to T . gondii ., Toll-like Receptor 2 signaling in CD4+ T cells is known to promote Th17 responses in vitro 29 and regulate the pathogenesis of autoimmunity in a model of experimental autoimmune encephalitis ( EAE ) ., During LCMV infection , IFNγ-producing CD8+ T cells ( Tc1 cells ) require intrinsic MyD88 signals for differentiation and survival 30 , 31 ., The importance of intrinsic MyD88 signals for the development of Tc17 cells that confer resistance against microbes including fungi remains poorly understood ., We have reported that IL-17-producing CD8+ T cells are indispensible in mediating vaccine immunity against fungal pneumonia in CD4+ T cell deficient mice 2 ., In the current study , we investigated the underlying mechanisms that enable the priming and development of these potent vaccine effectors ., Here , using a mouse model of vaccination against lethal fungal pneumonia caused by Blastomyces dermatitidis , we show that T cell-intrinsic MyD88 signals are required for Tc17 cell responses and immunity ., In contrast , Tc1 responses are relatively spared in the absence of such signals ., Unlike the situation for anti-viral CD8+ T cells , poor accumulation of anti-fungal Tc17 cells is not linked to accelerated death or reduced expression of anti-apoptotic molecules Bcl-2/Bcl-xL ., Instead , the poor accumulation is due to impaired proliferation that is mediated via Akt1 through the mTOR pathway ., Moreover , we show that IL-1R and TLR2 , and not IL-18R , are the upstream sensors and signaling receptors that initiate these anti-fungal Tc17 cell responses ., Thus , we describe the novel contribution of intrinsic MyD88 signals in Tc17 cells during the development of anti-fungal immunity , and the role of the AkT1-mTOR axis in fostering sustained proliferation of these cells and establishment of Tc17 memory and immunity in CD4+ T cell deficient hosts ., We initially investigated the general requirement of MyD88 signaling for Tc17 responses following fungal vaccination ., We adoptively transferred OT-I cells into naïve congenic wild-type and MyD88-/- mice , and vaccinated the animals with attenuated recombinant Blasomyces yeast expressing the OVA epitope SIINFEKL ., On day 18 post-vaccination , following ex vivo restimulation with anti-CD3/CD28 antibodies , we first analyzed the percentage and total number of endogenous Tc17 and Tc1 cells that lack MyD88 by gating on activated Thy1 . 2+ve CD8+ T cells ( CD44hi ) ( Fig 1A ) ., The endogenous , IL-17 producing CD8+ T cells in MyD88-/- mice were severely blunted in the draining lymph nodes ( dLNs ) and spleen , whereas IFN-γ producing cells were largely spared ( 8 . 8 fold vs . 2 . 2 fold reduction , respectively ) ., MyD88 signals therefore are required to promote the generation of Tc17 cell responses after fungal vaccination ., To dissect the intrinsic vs . extrinsic requirement for MyD88 , we analyzed the transferred , wild-type , OT-I cells bearing a distinct , congenic Thy1 . 1 marker ., Surprisingly , IL-17A+ and IFNγ+OT-I responses were largely intact in both the wild-type and MyD88-/- recipients ( Fig 1B ) ., Thus , intrinsic MyD88 signaling is involved in CD8+ T cell responses , especially for the Tc17 subset ., To study the intrinsic role of MyD88 in Tc17 cells , we pursued further approaches ., First , we purified CD8+ T cells from naïve wild-type and MyD88-/- mice and transferred them into naïve TCRα-/- mice ( S1A Fig ) ., Recipients were vaccinated and challenged by the pulmonary route to assess recall responses in the lung , which are reminiscent of vaccine responses in the dLNs ., Vaccinated TCRα-/- hosts that received wild-type CD8+ T-cells had pronounced Tc17 cell responses compared to unvaccinated recipients ( S1B and S1C Fig ) ., Vaccinated TCRα-/- hosts that received MyD88-/- CD8+ T-cells had significantly lower Tc17 responses vs . the recipients of wild-type cells ( ~10 fold , p≤0 . 05 ) ., These data support the hypothesis that intrinsic MyD88 signaling is required for Tc17 more than Tc1 responses to a fungal infection ( Fig 1A and 1B and S1 Fig ) ., In an alternative approach , we confirmed an intrinsic role of MyD88 for CD8+ T cell responses by using MyD88ΔT mice in which only T cells lack MyD88 ., After vaccination and analysis of the dLNs , Tc17 cells were significantly impaired in MyD88ΔT mice vs . wild-type mice , whereas Tc1 cells were relatively spared ( Fig 1C; p≤0 . 05 ) ., In yet another approach , to assess antigenic specificity and exclude possible developmental T-cell repertoire anomalies in MyD88ΔT mice , we tested OT-Imyd88-/- mice ., We transferred OT-I cells into congenic recipients , vaccinated with recombinant OVA yeast and analyzed SIINFEKL-specific Tc17 and Tc1 responses ., OT-I cells lacking MyD88 produced significantly less IL-17A compared to wild-type OT-I cells in the dLNs and spleen ( Fig 1D; p≤0 . 05 ) ., MyD88 signaling was relatively dispensable for IFN-γ responses ., In vitro studies with OT-I cells and OVA-vaccine yeast illustrated the non-redundant role of TCR signaling in fungal-induced Tc17 responses , and studies with naïve CD8+ T cells illustrated the cell intrinsic role of MyD88 for Tc17 cell responses ( S2A and S2B Fig ) ., Thus , intrinsic MyD88 signals preferentially affect Tc17 over Tc1 responses after fungal vaccination ., Previously , we showed that Tc17 cells were necessary for vaccine immunity in the absence of CD4+ T cells 2 ., Here , we explored the functional role of MyD88 signaling in vaccine resistance of CD4+ T cell depleted mice ., Unvaccinated wild-type mice failed to control pulmonary infection and harbored ~4 log cfu of yeast in their lungs , whereas vaccinated mice acquired sterilizing immunity ( Fig 2A ) ., Unvaccinated MyD88-/- mice have a slightly higher fungal burden than unvaccinated wild-type mice indicating MyD88 promotes innate resistance in the lung ., However , vaccinated MyD88-/- mice failed to acquire immunity and exhibited a fungal burden similar to unvaccinated wild-type mice ( Fig 2A ) ., Thus , MyD88 signaling is essential for vaccine immunity ., To assess a cell-intrinsic role of MyD88 for vaccine-induced CD8+ T cell immunity , we vaccinated MyD88ΔT mice ., Vaccinated MyD88ΔT mice had a significantly higher fungal burden than vaccinated control mice ( Fig 2B ) ., Vaccinated MyD88ΔT mice did have a lower fungal burden ( ~1 log ) than unvaccinated controls , suggesting contributions to vaccine resistance by IFN-γ , TNFα , GM-CSF and IL-17A that are MyD88 independent ., To correlate the resistance phenotype with cellular infiltration of cytokine producing CD8+ T cells , we harvested lungs 4 days after challenge ( peak of cell influx ) and analyzed cells by flow cytometry ., The percentage of IL-17A+ CD8+ T cells in the lungs was significantly lower in vaccinated MyD88ΔT mice than controls ( Fig 2C; p≤0 . 05 ) ., The total numbers of IL-17A+ , IFNγ+ and GM-CSF+ CD8+ T cells also were significantly lower in vaccinated MyD88ΔT mice than vaccinated controls , with a greater impact on Tc17 cells than Tc1 cells ( Fig 2D; ~8-fold vs . 2-fold , respectively ) ., Thus , impaired immunity in vaccinated MyD88ΔT mice was correlated with poor influx and/or accumulation of cytokine-producing CD8+ T cells in lungs , reflecting impaired vaccine responses in dLNs and spleens ( Figs 1C and 2 ) ., Collectively , these data suggest that intrinsic expression of MyD88 is required for vaccine-induced CD8+ T cell immunity and protective Tc17 cell responses ., In Fig 1 , we investigated the intrinsic role of MyD88 signaling by analyzing CD8+ T cell responses approximately 3 weeks after vaccination ., Here , we asked whether intrinsic MyD88 signaling affects CD8+ T cell responses during the early or late stages of expansion ., These phases of expansion include priming , differentiation and proliferation of effector CD8+ T cells ., To analyze the kinetics of CD8+ T cell responses during these phases , we vaccinated mice and assessed responses in the dLNs and spleens on days 0 ( naïve ) , 10 , 15 and 23 ., As early as day 10 , the percentage and total numbers of IL-17A+ CD8+ T cells were significantly blunted in the spleens of MyD88ΔT vs . wild-type mice ( Fig 3A and 3B ) ., These impairments were evident in the dLNs only later , by 15 to 23 days post-vaccination , suggesting that differentiated Tc17 cells become effector cells and emigrate from dLNs ., Tc1 cell responses also were reduced in the MyD88ΔT mice , but these impairments appeared later and were less pronounced than blunted Tc17 responses ., Of note , in the early stages of expansion , the activation ( CD44hi ) of total CD8+ T cells was less impaired in MyD88ΔT mice , suggesting that intrinsic MyD88 signaling preferentially affects Tc17 cell responses ( S3A Fig ) ., Tc17 cells produced significantly less IL-17 on a per cell basis in MyD88ΔT vs . wild-type mice ( mean fluorescence intensity: 9986±403 vs 6153±398; S3B Fig ) , suggesting that intrinsic MyD88 signaling shapes not only the quantity , but also the quality of Tc17 cells ., We stained for RORγt , but found an insignificant difference between the two groups ., Thus , MyD88 signaling in CD8+ T cells is required for optimal Tc17 cell expansion following fungal vaccination ., The net number of T cells during the expansion phase is governed by proliferation and apoptosis of effector cells ., Bcl-2 and Bcl-xL play an important role in survival of effector CD8+ T cells 32 ., We asked whether reduced expansion of Tc17 cells in MyD88ΔT mice is linked to the reduced expression of anti-apoptotic molecules Bcl-2 and Bcl-xL ., The expression levels of Bcl-2 and Bcl-xL in Tc17 ( and Tc1 ) cells were comparable in vaccinated wild-type and MyD88ΔT mice ( Fig 4A ) ., We also assessed active caspase 3 following ex vivo restimulation , and found no significant differences between the groups in either cytokine-producing cells or in total CD8+ T cells ( Fig 4B ) ., Finally , we stained effector CD8+ T cells with Annexin V to detect signs of early apoptosis and again found no difference ., Thus , reduced Tc17 cell responses in the absence of MyD88 signaling are not due to either reduced survival or augmented death of effector CD8+ T cells ., Our previous work showed that CD43 expression is higher in Tc17 than in Tc1 cells 2 ., CD43 signaling has a dichotomous role in effector CD8+ T cells; CD43 promotes expansion during the early phase of the T cell response , but augments apoptosis in the later phase 33 ., Similarly , CD27 signaling is necessary for survival and/or proliferation of effector CD8+ T cells 34 ., Therefore , we explored whether reduced accumulation of Tc17 cells induced by deficient MyD88 signaling was associated with decreased CD43 and CD27 expression ., Fig 4C shows the frequency of CD43+ and CD27+ expression on Tc17 and Tc1 cells ., As before , CD43 expression was higher on Tc17 than Tc1 cells , but there was no significant difference between cells from wild-type and MyD88ΔT mice ., Likewise , expression levels of CD27 on Tc17 and Tc1 cells were comparable between the groups ( Fig 4C ) ., Thus , poor Tc17 cell accumulation in the absence of MyD88 is neither due to augmented apoptosis nor to blunted CD43 or CD27 receptor expression ., We evaluated whether MyD88 signaling regulated the proliferation of effector CD8+ T cells during the expansion and contraction phases of the T cell response ., To evaluate CD8+ T cell proliferation , mice received BrdU for three intervals after vaccination ( Fig 5 ) ., BrdU+ Tc17 and Tc1 cells were analyzed at the end of each period ., Wild-type Tc17 cells in dLNs exhibited rapid proliferation by day 14 ( 80% ) , which peaked by day 21 ( ~93% ) and showed contraction or memory transition by day 30 ( ~82%; Fig 5 ) ., Similar results were found in spleens , especially at day 30 , where proliferation of Tc17 cells was dramatically reduced ( S4 Fig ) ., The proliferation of Tc17 cells in MyD88ΔT mice followed similar kinetics , however the percentage of BrdU+ cells was significantly lower on day 14 and remained lower at subsequent time points ., Unlike the proliferation defect in Tc17 cell , the absence of MyD88 signaling did not significantly affect Tc1 proliferation ( Fig 5 ) ., We also did not detect proliferation defects in MyD88-deficient , activated CD8+ ( CD44hi ) T cells during the early phases of expansion ., Thus , MyD88 signaling sustained the proliferation of Tc17 cells , but not Tc1 cells , throughout the expansion phase , without exhibiting delayed expansion following fungal vaccination ., mTOR has an important role in the metabolism and functions of both innate and adaptive immune cells 35 ., Under Th17 polarization conditions in vitro , rapamycin treatment inhibited mTOR , blunting the expression of IL-17a transcript and proliferation of CD4+ T cells 36 ., We postulated that mTOR mediates the proliferation and/or survival of Tc17 cells in MyD88-sufficient mice , and evaluated the effect of rapamycin treatment on Tc17 and Tc1 responses ., Rapamycin treatment significantly blunted the total number and percentage of vaccine-induced Tc17 cells in the dLNs and spleens of wild-type mice , but had no effect in MyD88ΔT mice ( Figs 6A and S5 ) ., The number of Tc17 cells was similar in rapamycin-treated wild-type mice and MyD88ΔT mice , supporting the requisite role of mTOR for MyD88-dependent Tc17 responses ., Rapamycin treatment inconsistently affected Tc1 cell responses in vaccinated wild-type mice , blunting the numbers of Tc1 cells in dLNs , but not in spleen , and leaving the percentage of Tc1 cells unaffected in these organs ., We next asked whether blunted Tc17 responses after rapamycin treatment are due to inhibition of proliferation ., Vaccinated mice were pulsed with BrdU and treated with either rapamycin or PBS control ., Treatment with rapamycin significantly reduced BrdU+ Tc17 cells in wild-type mice , but not MyD88ΔT mice ( Fig 6B ) ., In contrast , rapamycin treatment did not significantly affect the proliferation of Tc1 cells in either group of mice ( Fig 6B ) ., Thus , MyD88 signaling enhances antifungal Tc17 cell responses by augmenting proliferation of these cells via an mTOR dependent pathway ., Many kinases can phosphorylate and activate mTOR , including Akt 37 ., TLR2 ligation enhanced T-bet expression in CD8+ T cells and increased their cytotoxic functions , which were dependent on Akt and mTOR activation 38 ., As shown above , mTOR activity is likely modulated by MyD88 signaling ., Here , we asked if Akt signaling is required for MyD88-mediated Tc17 responses and mTOR phosphorylation ., We first assessed phosphorylation of Akt in CD8+ T cells stimulated by yeast in vitro ., We observed a pronounced increase in phosphorylation of Akt at T308 and S473 sites in the presence of DC supernatant from yeast-stimulated cultures ( S6A Fig ) ., We next assessed whether Akt1 is phosphorylated in a MyD88-dependent manner in CD8+ T cells activated in vitro ., We saw higher phosphorylation of Akt1 in wild-type CD8+ T cells compared to MyD88ΔT cells consistently at 20 , 40 and 60 minutes after activation ( S6B Fig ) ., To test the functional role of Akt signaling in vivo during vaccination , we inhibited Akt with compound A-443654 39 ., Akt inhibition blunted Tc17 responses in the dLNs and spleen of wild-type mice , but not MyD88ΔT mice ( Fig 7A ) ., Conversely , Akt inhibition did not affect Tc1 cell responses in wild-type mice ., Instead , Akt inhibition actually increased the numbers of IFN-γ producing cells in MyD88ΔT mice ., The effects of Akt inhibition resembled those of rapamycin treatment , suggesting a possible link between MyD88 activation of mTOR and Akt1 function in regulating downstream Tc17 responses ., To test a direct link between them , we analyzed the influence of Akt1 inhibitor on mTOR ( S2448 ) phosphorylation in CD8+ T cells ., Akt1 inhibited mTOR phosphorylation only in the presence of MyD88; that is , in wild-type CD8+ T cells , but in not MyD88ΔTcells , which confirmed that MyD88-dependent activation of mTOR occurred through Akt1 signaling ( Fig 7B ) ., Collectively , these results suggest that Akt1 signaling is required for MyD88 dependent Tc17 responses that are mediated through mTOR upon fungal antigen engagement and that these signals are promoted via IL-1 ( see section below ) ., We previously reported that Tc17 cells are reduced in IL-1R-/- mice 2 ., Others have documented TLR2 and IL-18R signaling in CD8+ T cells 38 , 40 , although the intrinsic role of IL-1R , TLR2 and IL-18R for Tc17 responses has not been described ., We assessed the function of these receptors in vitro and in vivo ., For in vitro studies , we purified CD8+ T cells and incubated them with wild-type or MyD88-/- BMDCs loaded with yeasts ., IL-17A levels were significantly lower in the supernatants from IL-1R1-/- , MyD88-/- , TLR2-/- and MyD88ΔT vs . wild-type CD8+ T cells , but were unaffected for IL-18R-/- CD8+ T cells ( Fig 8A ) ., The deficit in IL-1R-/- CD8+ T cells was similar to the MyD88-deficient groups , and more pronounced than for TLR2 deficient CD8+ T cells ., Phosphorylation of Akt at T308 and p-mTOR levels in CD8+ T cells were enhanced by the addition of either IL-1α or IL-1β or both , but only in the presence of MyD88 ( S6C Fig ) ., CD8+ T cells incubated with MyD88-/- BMDCs also produced significantly less IL-17A ( S7 Fig ) , which is consistent with an independent , extrinsic contribution ., For in vivo studies , we created mixed bone-marrow chimeras using irradiated TCRα-/- mice as a recipient for different donors ( Fig 8B ) , or administered blocking antibody against IL-18R throughout the study ., Tc17 and Tc1 cells were both reduced in the dLNs in the absence of T-cell specific MyD88 , IL-1R and TLR2 ( Fig 8C ) , but Tc17 cells were unaffected by blockade of IL-18R ( Fig 8D ) ., The spleens of chimeric mice revealed more pronounced impairments in Tc17 vs . Tc1 cells ( S7 Fig ) , however TLR2-/- Tc17 cell numbers were not significantly different from wild-type ., Thus , IL-1R and TLR2 exert hierarchical contributions to intrinsic MyD88 signaling in Tc17 cells , with the former being most important , and IL-18R signals appears to be dispensable in this model of vaccine-induced anti-fungal Tc17 cells ., Th17 cell responses are essential for immunity against infections including those caused by fungi 9 ., AIDS and other immune compromising disorders are associated with increased rates of opportunistic fungal infections due to CD4+ T cell lymphopenia 41 ., Hence , uncovering residual protective immune cells against fungi is essential for vaccination of at risk individuals ., We previously showed that in the absence of CD4+ T-cell help protective anti-fungal CD8+ T cell responses are elicited and maintained as long-lasting memory cells ., Tc17 cells are indispensible for this vaccine-induced fungal immunity 2 ., The extrinsic cytokine signals required for differentiation of Tc17 ( and Th17 ) cells have been characterized , and include TGFβ , IL-6 , IL-21 , and IL-23 ., However , the role of T cell intrinsic signals including MyD88 and upstream TLRs and IL-1R family members for Tc17 ( and Tc1 ) responses during immunity to infection has been less clear ., Here , we document a requisite role for intrinsic MyD88 in Tc17 cell responses and fungal vaccine immunity ., We also show that under the same ‘inflammatory milieu’ , intrinsic MyD88 signals are indispensable for Tc17 cell responses , whereas Tc1 cells are less affected in the absence of these signals ., MyD88 deficiency enhances susceptibility to infections caused by viruses , bacteria , parasites and fungi , but its contribution to resistance varies depending upon the pathogen ., MyD88 is essential for innate immunity and resistance without affecting CD8+ T cell responses during Trypanosoma infection 42 ., In contrast , MyD88 has a cell-intrinsic role for CD8+ T cell responses during lymphocytic choriomeningitis and vaccinia infections , where Tc1 responses are compromised in its absence 31 , 43 , 44 ., During experimental toxoplasmosis , T-cell-intrinsic MyD88 deficiency severely affects Th1 responses and impairs resistance 28 ., Our study unveils a critical role for intrinsic MyD88 function in CD8+ T cells during vaccine immunity against fungal pneumonia; its absence leads to a profound deficit of Tc17 responses in the lung ., We explored mechanisms underpinning MyD88 dependent , intrinsic control of Tc17 responses ., After antigen engagement , T cell expansion is the net result of effector T cell proliferation and death ., Several modes of cell intrinsic MyD88 action are possible ., Lack of intrinsic MyD88 signaling during viral infection enhanced apoptosis of effector CD8+ T cells ( despite normal Bcl-xL expression ) without affecting proliferation 31 ., In contrast , intrinsic MyD88 was required to sustain proliferation of effector Tc1 cells in a model of protracted viral infection 30 ., Our findings suggest that intrinsic MyD88 is required for sustained proliferation of Tc17 cells , but not Tc1 cells ., Our studies also show that MyD88-/- CD8+ T cells were not prone to apoptosis and that both Tc17 and Tc1 cells displayed similar levels of active-caspase3 , Annexin V and Bcl-xL expression ., Bcl-2 levels were also not influenced by MyD88 expression in Tc17 cells , however Bcl-2 levels were lower in Tc17 cells than in Tc1 cells ., The relevance of this finding is unclear , but our prior work showed that Tc17 cells portend long-term memory and display stem-cell like features 2 ., Akt signaling is integral for T cell activation and expansion 45 ., T cell differentiation may involve combinatorial signals that naïve T cells receive under a ‘micro-inflammatory milieu’ , but the role of TCR signaling in regulating T cell responses via MyD88-Akt for Tc17 cell responses is poorly understood ., Intrinsic MyD88 signals are known to boost functional avidity of IFNγ+ CD8+ T cell responses during vaccinia vaccination by reducing the activation threshold 46 , whereas our data show that these signals augment Tc17 responses more than Tc1 responses ., We also found that Akt1 signals were critical for boosting Tc17 , but not Tc1 responses , suggesting the hypothesis that low avidity CD8+ T cells require MyD88 signals to augment Tc17 responses whereas high avidity Tc1 cells may not require augmented Akt signaling ., This idea is in line with data from Th17 cells where low-strength T cell activation promotes their phenotype 47 ., Our in vivo results support this premise since Tc1 cells were unimpaired or even augmented in the presence of Akt1 inhibitor ., mTOR , a key metabolic sensor , is chiefly activated by PI3K-Akt pathway in T cells 35 ., To our knowledge the role of MyD88 in Akt-mTOR regulation of Tc17 cell responses has not been defined , although ligation of TLR2 has been shown to enhance T-bet and Tc1 cell responses in a manner dependent on Akt and mTOR 38 ., mTOR activity has been linked to Th17 cell responses by enhancing HIF-1α expression , Stat3 phosphorylation , RORγt translocation , and cell proliferation 48 ., We show here that pharmacological inhibition of mTORC1 reduced Tc17 cell , but not Tc1 cell proliferation in a MyD88-dependent manner ., We also found that MyD88 deficiency did not affect RORγt levels , similar to a report on Th17 cell polarization 36 ., That study , which involved in vitro Th17 polarization , showed that IL-1 signaling was required for the expression of IL-23R and together they enhanced mTOR activity to promote Th17 cell responses ., Other cytokines including IL-6 , IL-21 and Il-23 also can augment the expression of IL-23R 49 ., Our studies suggest that MyD88 signals influence mTOR through Akt to enhance Tc17 cell responses , and that MyD88 signaling may function independent of IL-23 signaling through Akt ., Nevertheless , IL-23 signaling in Th17/Tc17 cells may enhance Akt-mTOR signaling via Jak2 50 ., Further studies are needed to address whether IL-23 integrates MyD88 signaling downstream of mTOR ., One possible mechanism is that MyD88-Akt-mTOR may bolster stat3 function 48 , 51 that is activated by IL-6 , IL-21 and IL-23 ., Rapamycin treatment can also affect innate immunity by mechanisms that may , in turn , affect Tc17 cell responses 52 ., Our in vivo work suggested that Rapamycin treatment of vaccinated mice chiefly and selectively affected intrinsic MyD88 signaling for Tc17 cell responses , in view of the insignificant effect on Tc17 responses in MyD88ΔT mice ., Accumulating evidence suggests that IL-1 is required for both systemic and mucosal Th17 responses 53 ., IL-1 has pleotropic effects on both innate and adaptive immunity and the lack of IL-1 enhances the susceptibility to bacterial , viral and fungal infections 54 ., IL-1R-/- mice are vulnerable to coccidioidomycosis and blastomycosis , and IL-1 administration was shown to enhance fungal vaccine immunity in a manner that required IL-17R signaling 51 , 55 ., Administration of IL-1 enhanced the expansion and function of CD8+ T cells 55 and IL-1R-/- mice had reduced CD8+ T cell responses , IFN-γ production and viral clearance 56 , although the cell-intrinsic role of IL-1R signaling for CD8+ T cell responses was not explored ., Our study here shows that intrinsic IL-1 signaling sharply affects Tc17 cell responses ., This differing impact of IL-1 on Tc17 vs . Tc1 responses in the two studies may be due to the model system where differentiation towards Tc1 cell responses is favored and exogenous IL-1 just augmented the responses by increasing T-bet and activating mTOR 38 ., Ligands for TLR2 influence the polarization of Th17 cells in vitro and development of EAE in mice 29 ., Among the many ligands known for TLR signaling , fungi display zymosan , phospholipomannan , O-linked mannans , and DNA , which are recognized by TLR2 , TLR4 and TLR9 , respectively ., While impaired MyD88 signaling reliably enhances susceptibility to numerous fungal infections , the absence of individual TLRs shows varying results , perhaps due to impaired IL-1R family signaling in MyD88-/- mice or due to compensation by other TLRs 9 ., While a T-cell intrinsic role was not explored , IL-1R but not TLR2 signaling was essential for Th17 responses during Coccidioides infection 51 ., In an in vitro model , a TLR2 agonist was shown to enhance T-bet expression in CD8+ T cells by activating Akt and mTOR 38 ., Here , we observed that intrinsic TLR2 signaling was essential for Tc17 as well as Tc1 responses ., Differences among these studies may be due to the model , fungal strain , T cell type or compensation by other receptors ., Our data also suggested that TLR2 was only essential for initial priming in the skin dLNs , but not in the spleen , where exuberant circulating IL-1 may compensate the defect for Tc17 responses ., Collectively , we show that intrinsic MyD88 signals are required for anti-fungal vaccine immune responses in vulnerable CD4+ T cell deficient hosts through sustained proliferation and preferential expansion of Tc17 cells , which is dependent on Akt and mTOR ., Our study therefore identifies unappreciated targets for augmenting adaptive immunity against pathogenic fungi ., Our findings are important for designing vaccines against fungal infections in at risk individuals with CD4+ T cell defects and for immunotherapeutic intervention during infection and possibly autoimmune disorders ., Animal procedures were done in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Instit

Introduction, Results, Discussion, Methods

Fungal infections have skyrocketed in immune-compromised patients lacking CD4+ T cells , underscoring the need for vaccine prevention ., An understanding of the elements that promote vaccine immunity in this setting is essential ., We previously demonstrated that vaccine-induced IL-17A+ CD8+ T cells ( Tc17 ) are required for resistance against lethal fungal pneumonia in CD4+ T cell-deficient hosts , whereas the individual type I cytokines IFN-γ , TNF-α and GM-CSF , are dispensable ., Here , we report that T cell-intrinsic MyD88 signals are crucial for these Tc17 cell responses and vaccine immunity against lethal fungal pneumonia in mice ., In contrast , IFN-γ+ CD8+ cell ( Tc1 ) responses are largely normal in the absence of intrinsic MyD88 signaling in CD8+ T cells ., The poor accumulation of MyD88-deficient Tc17 cells was not linked to an early onset of contraction , nor to accelerated cell death or diminished expression of anti-apoptotic molecules Bcl-2 or Bcl-xL ., Instead , intrinsic MyD88 was required to sustain the proliferation of Tc17 cells through the activation of mTOR via Akt1 ., Moreover , intrinsic IL-1R and TLR2 , but not IL-18R , were required for MyD88 dependent Tc17 responses ., Our data identify unappreciated targets for augmenting adaptive immunity against fungi ., Our findings have implications for designing fungal vaccines and immune-based therapies in immune-compromised patients .

Patients with AIDS , cancer or immune suppressive treatments are vulnerable to infection with invasive fungi ., We have found that even when helper CD4 T cells are profoundly reduced in a mouse model that mimics this defect in AIDS , other remaining T cells are capable of mounting vaccine immunity against a deadly fungal infection , and they do so by producing the powerful , soluble product , IL-17 ., It has been widely believed that the activation and instruction of such cells , called Tc17 cells , is governed by another population of immune cells in the body , but we have found here that pathways within these Tc17 cells themselves mediate their activation and ability to produce the IL-17 needed for resistance to infection ., We have also identified elements of the circuitry controlling this pathway—elements called MyD88 , Akt1 and mTOR—and found that they control the production of IL-17 and not other products such as IFN-γ often produced by these cells ., Further , we determined that this circuitry controls the development of Tc17 cells by regulating their ability to divide and expand ., Thus , in a mouse model of vaccination against lethal fungal pneumonia caused by Blastomyces dermatitidis , we uncovered an important cellular arsenal that can be recruited to bolster resistance against a fungal infection , and identified novel ways in which the cells develop and expand into potent killers of fungi .

journal.pntd.0003595

Modeling the Dynamics of Plasmodium vivax Infection and Hypnozoite Reactivation In Vivo

Plasmodium vivax is one of the major agents of malaria infection , with around 2 . 5 billion people living in areas at risk of infection , and more than 70 million estimated annual infections 1–3 ., P . vivax is generally less pathogenic than Plasmodium falciparum infection due to the absence of sequestration and cytoadherence , and all blood-stage forms can be detected in peripheral circulation4 ., P . vivax also differs due to the development of dormant hypnozoite forms in the liver , which serve as reservoir of infection after the clearance or treatment of the acute blood stage of infection ., P . vivax shows a preference for reticulocytes as its host cell in the blood-stage of its life-cycle 5 ., The potential for reactivation of dormant hypnozoites creates a number of difficulties in understanding the transmission dynamics of P . vivax infection6 , however the majority of diagnosed infections are thought to be due to hypnozoite reactivation rather than new primary infection 7–9 ., After treatment to eliminate blood stage infection , new infections may occur as a result of recrudescence ( failure of treatment of the blood stage ) , hypnozoite reactivation , or new primary infection ., Although comparison of P . vivax genotypes may be useful in distinguishing recrudescence of the blood stage parasites after treatment , it is not always useful in differentiating reactivation of a dormant hypnozoite from new primary infection 10–12 ., This is because the parasites causing relapses are often genetically different from those observed in the most recent blood-stage infection , so it is not possible to differentiate reactivation from new primary infection using genotyping13 , 14 ., Therefore , it is difficult to know the proportion of blood stage infections due to hypnozoite reactivation versus new primary infection by P . vivax ., In this study , we aim to apply mathematical modeling to quantify the relative contribution of new primary infection , and infection initiated due to the reactivation of dormant liver stage hypnozoites ., We analyse data from two published prospective studies where individuals were treated to eliminate blood-stage infection , and a subset were treated with primaquine , a licensed radical treatment for hypnozoites , to eliminate preexisting hypnozoites 15 , 16 ., By comparing the rate of observed blood stage infection in the two groups , we can estimate the contribution of primary infection versus hypnozoite reactivation in P . vivax infection ., The field study data were from a published prospective study with recruitment from July 1995 to July 1996 , on 342 individuals of different ages ( 68% ( 258 ) <15 years of age ) living on the western border of Thailand where P . vivax is endemic9 ., Individuals with asexual forms of P . vivax on a blood smear were enrolled , treated with chloroquine ( 25mg base/kg over 3 days ) and followed up until presentation of pure or mixed P . vivax blood stage infection ., Individuals with reappearance of pure P . vivax infection were retreated with either chloroquine only ( 70 individuals with mean age ( range ) 12 ( 1–50 ) ) or chloroquine and primaquine ( 0 . 25 mg/kg daily for 14 days , 43 individuals with mean age ( range ) 13 ( 5–43 ) ) , and were followed up by microscopy until detection of P . vivax blood stage parasites ., Each dose of chloroquine was supervised and the patient observed for 1 hour after dosing ., The criteria for enrollment and method of detection and quantification of P . vivax parasites in the blood smears are detailed elsewhere9 ., From the fact that primaquine can kill liver stage hypnozoites15 , 16 we classified the data in two groups: individuals retreated with chloroquine+primaquine ( CQ+PQ ) group and individuals retreated with chloroquine only ( CQ only ) group ., We also analysed published data on a treatment-time-to-infection study in PNG 8 ., In this study , the contribution of relapse to the risk P . vivax infection and disease was studied in 433 PNG children 1–5 years of age ., Children were randomized into one of three groups: ( 1 ) artesunate ( 4mg/kg/d for 7 days ) plus primaquine ( 0 . 5 mg/kg/d for 14 days ) , 149 individuals , ( 2 ) artesunate only ( 4 mg/kg/d for 7 days ) , 150 individuals or ( 3 ) no treatment ( control ) , 150 individuals , and were followed up for infection and the presence of febrile illness ., Every dose of treatment was administered as direct observed therapy ., The criteria for enrollment , method of randomization , treatment procedures and method for the detection and quantification of P . vivax parasites in the blood smears were detailed in the original publication 8 ., For our analysis we extracted data on infection rates from Fig . 2 of Betuela et al 8 , using Grafula 3 . 0 ( Knowledge Probe Inc , Aurora ) to extract data on cumulative proportion of infections detected by light microscopy ., The data was classified into two groups; those receiving artesunate+primaquine ( AS+PQ group ) , and artesunate only ( AS only ) group ) ., The dynamics of P . vivax infection are characterized by both primary infection and reactivation ., For individuals living in P . vivax endemic regions both primary infection and activation of hypnozoites occur throughout the year ., For modelling purposes we will label the two groups of subjects as, i ) the B+H group , comprising individuals who received drugs against both blood stage parasites and hypnozoites and, ii ) the B group , comprising individuals who received drugs against blood stage parasites only ., We initially assume 100% drug efficacy against both the blood and liver stage parasites ., Thus for individuals receiving in the B+H group , all observed infection is due to new primary infection ., For individuals in the B group , infections can arise either from new primary infection , or from reactivation of hypnozoites ., Thus , we can model the time to first P . vivax infection after treatment , and fit this to the ‘survival curve’ of time to detection of P . vivax infection after treatment ., The proportion of treated individuals in the B+H group remaining uninfected at a given time t is indicated as ( SB+H ( t ) ) and follows the equation:, SB+H ( t ) ={1 , t≤d1e−k ( t−d1 ) , t>d1 ,, ( 1 ), and the proportion of treated individuals in the B group remaining uninfected at time t is indicated as ( SB ( t ) ) and follows the equation, SB ( t ) ={1 , t≤d2e−k ( 1+c ) ( t−d2 ) , t>d2, ( 2 ), Here k is the rate of initiation of new primary infections , c is the relapse to reinfection ratio and d1 and d2 are delays to detection of blood stage P . vivax parasites in the groups respectively ., Fig . 1A and 1B show the schematic representation of Equation ( 1 ) and ( 2 ) where individual received treatments for either both blood stage and liver stage parasites or blood stage parasites only ., The modeling above assumes that all primaquine is completely effective in all individuals ., However , if primaquine fails to kill hypnozoites from a proportion of strains , or in a proportion of individuals , then we expect altered dynamics ., If primaquine kills only a fraction of hypnozoites , then the equation for SB+H ( t ) becomes:, sB+H ( t ) ={1 , t≤d1e−k ( 1+Rc ) ( t−d2 ) , t>d1, ( 3 ), Where R is the fraction of hypnozoites that are resistant to primaquine ., We note that in our study we cannot differentiate between this and Equation ( 2 ) with a different c ., If primaquine only kills hypnozoites in a subset of individuals , then primaquine resistant individuals will behave as if they have not received primaquine ., Thus , if primaquine is only effective in a proportion of individuals , the rate of infection of individuals in the B+H group will be;, sB+H ( t ) ={1 , t≤d1Pe−k ( t−d1 ) + ( 1−P ) e−k ( 1+c ) ( t−d1 ) , t>d1 ,, ( 4 ), where P is the proportion of individuals in whom primaquine is effective ., The dynamics is illustrated in Fig . 1C ., Equations ( 1 ) , ( 2 ) and ( 4 ) were fit to the data using the lsqnonlin function in MATLAB R2012 ( Release M ( 2012 ) The MathWorks Inc , Natick , MA , USA ) , which uses a nonlinear least-squares method , to understand the contribution of hypnozoites reactivation to the dynamics of P . vivax infection ., To understand the different contributions of primary infection and reactivation , we need to understand how the force of infection from primary infection and hypnozoite reactivation evolve over time / with exposure starting with a naïve host ., If we consider a fixed rate of infectious mosquito inoculation ( M ) , and that each inoculation infects a number of liver cells ( S ) that will eventually go on to produce a blood stage infection , then the total rate of successful infection of liver cells is simply MS . Assuming that the time spent in the liver stage is negligible for primary infection , the rate of new primary infections ( Ip ) is simply the overall rate of infection of liver cells MS multiplied by the fraction of liver cell infections that result in primary infection ( f ) :, IP=fMS, ( 5 ), We note that fMS is equivalent to k in Equations 1–4 ., Importantly , the rate of successful infection of liver cells may not be the same as the rate of successful mosquito inoculation ( as one successful inoculation may infect one or more liver cells , see Fig . 1 ) ., Moreover , the definition of successful infection of a liver cell is not simply that a cell becomes infected , but that the infected cell gives rise to a blood stage infection at some stage ( either immediately or with some delay ) ., If we consider an individual that starts with no hypnozoites ( for example a neonate , or someone successfully treated with primaquine ) , the number of hypnozoites and rate of infection from hypnozoite reactivation ( IH ) evolves over time as hypnozoites accumulate upon repeated infection ., Hypnozoites accumulate dependent on the rate of infection of liver cells ( MS ) , and the fraction of liver cells that become hypnozoites ( 1-f ) ., The dynamical equation for the number of hypnozoites ( H ) over time is, dHdt= ( 1−f ) MS−aH, ( 6 ), where a is the reactivation rate of hypnozoites ., We note that this assumes that hypnozoites never infect more than a tiny proportion of total liver cells ., The rate of infection due to hypnozoite reactivation is then simply:, IH=aH, ( 7 ), The solution of Equations ( 6 ) and ( 7 ) is, IH= ( 1−f ) MS ( 1−e−at ) +aH0e−at, ( 8 ), where H0 is the initial number of hypnozoites ., At steady state , the rate of infection from hypnozoites ( IH ) in Equations 7 and 8 is equivalent to kc in Equations 2 and 4 ., Similarly , the total rate of new blood stage infections at steady state is simply MS ( the rate of successful infection of liver cells ) ., Fig . 2 illustrates the mechanisms of infection and the relationship between parameters in Equations 1–4 , and 5–7 ., Plasmodium vivax infection often follows a seasonal pattern , with higher infection in the wet season17 , 18 ., To understand the role of fluctuations in the force of infection and the effect of seasonality , we modified Equation ( 6 ) by allowing the rate of infectious mosquito inoculation to vary seasonally;, M=Ms ( 1+Mfcos\u2061 ( 2πt/365 ) ), ( 9 ), where Ms is the mean rate of infectious mosquito inoculation and Mf is a parameter for the degree of seasonal fluctuation ., The periodicities of these functions are such they divide the season into dry and wet seasons ., We examine the effect of seasonality and EIR on the contribution of primary infection to reactivation by numerical simulations of Equations ( 5 ) and ( 6 ) with seasonality terms included ., We first estimated using Equation ( 1 ) the rate of infection in individuals receiving chloroquine plus primaquine treatment in Thailand , in whom all infections are assumed to be due to new primary infection ., Assuming a constant force of infection , we found that a rate of infection of approximately 0 . 0017 per day ( equating to an average time to primary infection of 588 days ) provided the best fit to the data ( Fig . 3A ) ., We next estimated the rate of infection and reactivation occurring in the individuals that received chloroquine alone using Equation ( 2 ) , where infection can arise due to both new primary infection as well as hypnozoite reactivation ., In this case , we observed the rate of 0 . 043 infections per day ( equivalent to an average of 23 days to infection ) ., Since the individuals receiving chloroquine alone experience both primary infection and reactivation from hypnozoites , we can subtract the rate of primary infection ( estimated in the CQ + PQ group ) to estimate the rate of hypnozoite reactivation ., Thus , we estimate a rate of hypnozoite reactivation of 0 . 0413 per day ( equivalent to a hypnozoite reactivation every 24 days ) ., There was no evidence for a difference in the delays from treatment to first detection in the CQ+PQ group and the CQ group ( p = 0 . 8563 , F-test ) ., The relative rates of new primary infection and hypnozoite reactivation estimated in this data suggest that approximately 4% of infection events in individuals receiving chloroquine occurred due to primary infection , and approximately 96% of infection events occurred due to hypnozoite reactivation ., This implies a ratio of primary infection to hypnozoite reactivation of approximately 1 to 24 ., Once hypnozoites are laid down , they will reactivate at some later time either spontaneously , or following some form of stimulation ( such as fever or concurrent infection ) ., So , for a single bite we might imagine there is some average time to reactivation , and a distribution in the probability of reactivation with time ., The delay between initial inoculation and subsequent hypnozoite reactivation is highly variable19 , 20 ., However , even in the absence of knowing the precise schedule of reactivation of individual hypnozoites , we can still understand the dynamics of reactivation in an endemic setting ., That is , in an endemic setting we do not observe reactivation from a single inoculation , but in fact from a long series of past inoculations ., We have a probability of reactivation from an inoculation 6 months ago , which is dependent on the rate of inoculation six months ago , the proportion of hypnozoites surviving 6 months , and the rate of reactivation of 6 month old hypnozoites ., The same is true for hypnozoites inoculated a month ago , or a year ago ., In this circumstance if we have a constant rate of inoculation for individuals of a particular age group who have been exposed to Pv infections for a sufficient time to reach ‘steady state’ of infection , where each the average rate of reactivation of hypnozoites reflects the rate at which they were laid down ( from Equation ( 6 ) , when the number of hypnozoites is constant, ( dHdt=0 ), , then the rate of hypnozoite reactivation ( aH ) equals the rate of new hypnozoite infection ( ( 1-f ) MS ) ., Since treatment for blood-stage infection is relatively short-lived compared to the history of infection ( and has no impact on accumulated liver stages ) , this should not significantly affect this steady state ., Although we can estimate the ratio between primary infections and reactivations , we cannot directly estimate the ‘rate of infection’ ( rate of new infectious inoculation from mosquito bites ) from this data ., That is , for an inoculation to be infectious , it must eventually produce either primary infection , or a hypnozoite that later reactivates ., It is not clear that all inoculations must produce a primary infection ( some may produce only hypnozoites ) ., Thus , the rate of new primary infection in CQ+PQ individuals may or may not reflect infectious inoculation rate , since some fraction of inoculations may produce only hypnozoites ., However , the minimal rate of infectious inoculation would occur when every new inoculation produced a primary infection ., If we assume that every new inoculation must result in an early , acute blood stage infection , then the minimal inoculation rate is simply the rate of new blood stage infection in the CQ + PQ group ( and both inoculation and new primary infection would be experienced approximately every 588 days ) ., If this were the case , it would also require that each new inoculation must lay down approximately 24 hypnozoites ( in order to account for the observed high rate of hypnozoite reactivation in the CQ group ) ( Fig . 3B ) ., The alternative scenario occurs if some infectious inoculations do not produce a primary infection , and instead only lay down hypnozoites ., Since to be an ‘infectious inoculation’ the inoculation must produce either one primary infection or one hypnozoite , the highest rate of infectious inoculation would be when each infection only produced exactly one infected liver cell , which could either produce one primary infection or one hypnozoite ., In this case , the rate of infection observed in the CQ group is exactly the rate of infectious inoculation ., If this were the case , then it would require that only 4% of infectious inocula presented as a primary infection , and the rest became dormant ( laid down their one hypnozoite ) without being observed as a primary infection ( Fig . 3C ) ., This seems unlikely , as there is experimental evidence that more than one reactivation event ( and thus more than one hypnozoite ) can arise from a single inoculation21 ., The analysis above describes the maximal and minimal rates of infectious inoculation , which imply very different numbers of liver cells infected per inoculation ., The maximal rate suggests that 24 liver cells are infected from each infectious bite , but that this only occurs every 588 days ., The minimal rate suggests that each infectious bite infects at most one liver cell , but this happens as frequently as every 23 days ., The rate of infection could also be considered in terms of the rate of infection of liver cells ( either as primary infection or hynozoites ) , even without knowing how many liver cells are infected per mosquito inoculation ., The infection rate in the CQ treated group is driven by both the rate of primary infection plus the rate of hypnozoite reactivation ., As discussed above , the rate of reactivation reflects the sum of all previous inoculations ( at all times ) and their probability of reactivating , and is thus reflective of the rate of ‘laying down’ hypnozoites ., The infection rate of the CQ group is thus the total rate of infection of liver cells ( either destined for primary infection or hypnozoites ) , assuming that only one cell initiates each primary infection or reactivation ( assuming chloroquine is effective ) ., Thus , the minimum rate of liver cell infection can be derived from the CQ group , and is 0 . 043 infected cells per day ( a new infected liver cell every 23 days ) ., However , unless we know the number of liver cells infected on each inoculation , or the proportion of inoculations that cause primary infection , we cannot directly estimate the inoculation rate ., From the analysis above we can estimate the minimal and maximal inoculation rates , and the minimum rate of production of new infected liver cells ( which is the same as the maximal inoculation rate ) ., Fig . 3D illustrates one of many possible scenarios in between these extremes ., For example , if we had an infection rate of 0 . 0086 per day ( equating to a new infection being established every 120 days ) , this would require that approximately 20% of infection were observed as a primary infection , and each infection produced on average five hypnozoites ( Fig . 3D ) ., However , it is clear that although we can estimate primary infection and reactivation rates , we cannot directly estimate the rate of infectious inoculation from the data unless we assume that each infection event always produced an observed primary infection ., In addition , we cannot estimate the rate of infection of liver cells unless we assume that each infection arises from a single infected liver cell ., In order to compare these dynamics in another population , we investigated the infection rates of patients treated with either artesunate alone , or artesunate + primaquine by analyzing a published data set from Papua New Guinea ., In order to make our analysis directly comparable with the Thai study , we first restricted our analysis to infections detected by microscopy in the first 60 days since treatment ( Fig . 4A ) ., In individuals receiving artesunate plus primaquine , we observed using Equation ( 1 ) a rate of blood stage infection of 0 . 0102 / day ( equivalent to 98 days between new blood stage infections ) ., In individuals receiving artesunate alone , we observed using Equation ( 2 ) a rate of blood stage infection of 0 . 0344 / day ( equivalent to a new infection every 29 days ) ., Using the same approach as discussed above , we would estimate that in individuals treated with artesunate alone , 30% of infections occur due to primary infection and 70% due to hypnozoites reactivation ( a ratio of 2 . 37 to 1 ) ., This ratio of reactivation to primary infection is 10 fold lower than the ratio observed in the Thai study ., There was no evidence for a significant difference in the estimated delays to first detection of infection in AS+PQ group and AS only groups ( p = 0 . 0807 , F-test ) ., The large difference in the ratio of primary infection to hypnozoite reactivation between the Thai and PNG studies raises a number of questions , which we explore below ., One reason for the difference between the Thai and PNG studies could arise from primaquine resistance ., That is , either particular strains of parasites or particular individuals may be resistant to primaquine ., If primaquine were only effective in a subset of parasite strains , then only a fraction of hypnozoites ( from sensitive strains ) would be killed by treatment ., Hypnozoite reactivation would then be reduced by this fraction , and we would see simply an increase in the ( exponential ) rate of infection in the primaquine treated group ., Alternatively , if primaquine was only effective in a proportion of individuals , then we might expect to see a rapid rate of infection in primaquine-resistant individuals ( at the same rate as those receiving artesunate alone , due to both primary infection and hypnozoite reactivation ) , and a slow rate of primary infection in those in whom primaquine was effective ., This would produce a different ( non-exponential ) infection curve , characterized by two populations and two infection rates ., In order to assess whether primaquine resistance might either affect a proportion of strains , or a proportion of individuals , we modeled the full time course of infection in the PNG data ( comparing Equations ( 1 ) , ( 2 ) and ( 4 ) ) ., We found that a model in which a proportion of individuals are primaquine-resistant ( Equation 4 ) provides a significantly better fit to the data ( p<0 . 0001 , F-test ) ., In the case of primaquine resistant individuals , the infection rate in those in whom primaquine was ineffective should be the same as the infection rate in those receiving artesunate alone , and we can estimate the rate for those with successful primaquine therapy independently ., When we fit the data to this function ( Equation 4 ) , we can estimate that the best fit to the data occurs if primaquine is effective in ≈60% percent of individuals , and the rate of primary infection ( in the group in which primaquine was effective ) was 0 . 0032 / day ( equivalent to a new primary infection every 313 days ) ., Interestingly , this gives a ratio of primary infection to reactivation of 1 to 9 , which is much more similar to the rate estimated from the Thai study ., Table 1 shows the best-fit estimates of the models ( 1 ) , ( 2 ) and ( 4 ) to the both Thai and PNG data sets ., A second difference between the Thai and PNG studies is the age of the cohorts , which was young in PNG , but included all ages in Thailand ., Therefore we asked whether age may affect the proportion of infections arising from hypnozoite reactivation ., Primary infection arises soon after infectious inoculation , and thus should be proportional to the current rate of infectious inoculation ., By contrast , reactivation from hypnozoites requires first the establishment of a ‘reservoir’ of hypnozoites from previous infections , and then their later reactivation ., Thus , for example , after the first exposure in life , only hynozoites laid down by the first inoculation can reactivate ., However , after many exposures , reactivation can occur from hynozoites laid down at different times in the past ., This is evident from previous studies of P . vivax clonotypes in infection ., These studies have analysed the relationship between P . vivax clonotypes from baseline infection , and in subsequent infectious episodes ., In adults , these clonotypes are very often unrelated , consistent with reactivation of hypnozoites laid down prior to the most recent infection event 13 , 14 , 22 ., However , in children it is more likely that baseline infection and subsequent reactivation will be due to the same clonotype 23 ., We modeled the rates of primary infection and reactivation with age , to investigate how this might impact our analysis ( using Equations ( 5 ) , ( 6 ) and ( 7 ) ., The results suggest that for a given infection rate , the rate of primary infection is constant over time , but the rate of reactivation takes some time to reach its long-term level ., Therefore at young ages we would expect an overall lower rate of infection ( due to a lower rate of hypnozoite reactivation ) , as well as a higher ratio of primary infection to reactivation ., How long this effect would be observed is directly related to the average time between laying down of hypnozoites and their subsequent reactivation ( as shown in Fig . 5A ) ., However , since the average time for hypnozoites to reactivate is thought to be of the order of months in tropical regions , this effect should only be present very early after initial exposure , and seems unlikely to have been the cause of the observed differences between the PNG and Thai cohorts ., The analysis above suggest that if age plays an important role in the ratio of primary infection to reactivation , then younger children treated with CQ in the Thai cohort should also exhibit an overall lower infection rate than older individuals ( because of the reduced rate of hypnozoite reactivation ) ., To test this we analysed the rate of infection of the children aged <5 years in the CQ treated cohort from the Thai study ., Our fitting using Equation ( 2 ) showed no evidence for a significant difference in the rate of infection between those aged < 5 years or >5 years old in the Thai cohort ( p = 0 . 2493 , F-test , Fig . 5B ) ., Thus , it seems unlikely that age alone would explain the difference between the Thai and PNG studies ., The model above considers how the infection rate from new inoculation and reactivation of P . vivax hypnozoites evolves with age in children exposed to infection ., The same dynamics of accumulation of hypnozoites will also occur after successful primaquine therapy , when again the individual starts with no hypnozoite reservoir ., After successful primaquine therapy both the number of hypnozoites and the rate of infection from hypnozoites will similarly increase over time ., Other mechanisms for the differences between the Thai and PNG studies include the overall rate of inoculation , as well as seasonal fluctuations in this ., Changes in the inoculation rate per se should not affect the ratio of primary infection to reactivation ., That is , the average ratio of primary infection to reactivation is determined by the proportion of sporozoites progressing immediately to infection versus becoming hypnozoites , and is relatively independent of the rate of inoculation ., Moreover , comparing the groups with treatment for blood stage infection only ( CQ in Thailand and AS in PNG ) , the rate of infection in these groups was similar ( 0 . 043 vs . 0 . 034 infections per day , respectively ) ., Once we accounted for treatment resistance , our estimated rate of new primary infection was also similar ( 0 . 0017 vs . 0 . 0032 per day , respectively ) ., Thus , it seems unlikely that differences in the rate of infection were a major factor ., Seasonal fluctuations in the rate of new infections will have a direct effect on the rate of primary infection over time ., However , if the reactivation of hypnozoites happens on a longer timescale , it may be less susceptible to seasonal variation , and thus alter the ratio of primary infection to hypnozoite reactivation over the seasons ., Since the PNG study may have a higher degree of seasonality compared with the Thai study 8 , 9 , we explored the predicted impact of seasonality ., We modeled a sinusoidal variation in the rate of overall infection ( Equation ( 9 ) ) , and a constant fraction of infections becoming primary infection ( Equation ( 5 ) or being laid down as hypnozoites ( Equation ( 6 ) ) and later reactivating ( Equation ( 7 ) ) ., The rate of infection from primary infection and hypnozoite reactivation thus evolve over time as described in Equations ( 5 ) and ( 8 ) respectively ., When the average time to hypnozoite reactivation was short ( one month , Fig . 6A ) , the number of hypnozoites and rate of hypnozoite reactivation fluctuates a lot over time , mirroring ( with slight delay ) the fluctuations in primary infection ., However , as the average time to hypnozoite reactivation gets longer ( Fig . 6B and 6C ) ) , the number of hypnozoites and rate of hypnozoite reactivation becomes less variable with season ., For our purposes , we are most interested in how seasonal fluctuation in infection rate might affect the ratio of primary infection to hypnozoite reactivation ., Somewhat counter-intuitively , the shorter the time to hypnozoite reactivation , the less seasonal fluctuation in this ratio is seen ( Fig . 6D ) ., Thus , seasonality of infection may significantly alter the ratio , but this is least likely to have an effect with tropical strains of P . vivax , where the time to hypnozoite reactivation is thought to be relatively short ., In our modeling , the rate of hypnozoite reactivation was assumed to be independent of season ., However some have suggested that there may be a seasonality in presentation from UK residents returning from endemic regions24 , supporting a varying reactivation rate with season , which could further affect the ratio of primary infections to reactivation ., In the study we have estimated the relative contribution of new primary infection and the reactivation of dormant liver-stage hypnozoites to the rate of blood stage infection in individuals living in the P . vivax endemic areas ., Our modeling results showed that the vast majority of infections ( 96% ) in Thailand were due to hypnozoite reactivation ., The proportion of infection due to hypnozoite reactivation in PNG was less clear ., Considering only the early phase of infection and 100% efficacy of primaquine , we would estimate that only 70% of infections in PNG were due to hypnozoite reactivation ., However , we found that when the full time course of infection was considered , we had a significantly better fit to a model in which up to 40% of PNG individuals were resistant to primaquine therapy ., If this level of primaquine resistance in the population is correct , then 90% of the infections were due to hypnozoite reactivation in the PNG cohort study ., A number of factors differed between the Thai and PNG studies that might affect our results ., Firstly , the age of the cohorts was different , with the PNG cohort focused on children 1–5 years of age , whereas the Thai cohort included individuals of all ages ., Age may play a role in susceptibility to P . vivax infection 25 , 26 , and in the ratio of primary infection to reactivation ., However , if we restricted our analysis to the 1–5 years age group in the Thai study , we found a similar overall infection rate to the rest of the cohort ., Secondly , th

Introduction, Materials and Methods, Results, Discussion

The dynamics of Plasmodium vivax infection is characterized by reactivation of hypnozoites at varying time intervals ., The relative contribution of new P . vivax infection and reactivation of dormant liver stage hypnozoites to initiation of blood stage infection is unclear ., In this study , we investigate the contribution of new inoculations of P . vivax sporozoites to primary infection versus reactivation of hypnozoites by modeling the dynamics of P . vivax infection in Thailand in patients receiving treatment for either blood stage infection alone ( chloroquine ) , or the blood and liver stages of infection ( chloroquine + primaquine ) ., In addition , we also analysed rates of infection in a study in Papua New Guinea ( PNG ) where patients were treated with either artesunate , or artesunate + primaquine ., Our results show that up to 96% of the P . vivax infection is due to hypnozoite reactivation in individuals living in endemic areas in Thailand ., Similar analysis revealed the around 70% of infections in the PNG cohort were due to hypnozoite reactivation ., We show how the age of the cohort , primaquine drug failure , and seasonality may affect estimates of the ratio of primary P . vivax infection to hypnozoite reactivation ., Modeling of P . vivax primary infection and hypnozoite reactivation provides important insights into infection dynamics , and suggests that 90–96% of blood stage infections arise from hypnozoite reactivation ., Major differences in infection kinetics between Thailand and PNG suggest the likelihood of drug failure in PNG .

Plasmodium vivax is one of two major parasite species causing human disease ., This parasite can lie dormant in the liver as a hypnozoite , before later reactivating to cause blood-stage infection ., Treatment to eliminate the dormant hypnozoite stage relies mostly on a single drug—primaquine ., Understanding the rate of primary infection versus hypnozoite reactivation is important to understanding primaquine efficacy and drug resistance , as well as the development of new drugs targeting hypnozoites ., Here we use mathematical modeling to analyse data from two clinical cohorts and show that up to 96% of infections may be caused by hypnozoite reactivation ., We also use modeling to understand the impact of drug resistance , seasonal infection and subject age .

journal.pgen.1006547

A Signature of Genomic Instability Resulting from Deficient Replication Licensing

Licensing of DNA replication origins begins early during the G1-phase of the cell cycle when ORC and CDC6 recruit CDT1-MCM2-7 to the chromatin reviewed in 1 ., MCM2-7 is a heterotypic hexameric complex that is the replicative helicase and , once loaded , its association with the DNA is stable until it becomes active during S-phase ., Licensing is restricted to the G1-phase of the cell cycle to prevent endo-reduplication of the DNA by relicensing already duplicated DNA during S-phase ., Many more sites are licensed with MCM2-7 complexes than are typically used in S-phase and these dormant origins are thought to serve as a backup system for recovering replication in the event of replication fork stalling or collapse 2–3 ., The importance of this system of backup origins for maintaining genome stability is demonstrated by the observation that mice in which MCM2-7 proteins are deficient or compromised develop normally but exhibit increased genome instability , high rates of cancer and stem cell deficiencies 4–6 ., The cancers that arise in MCM deficient mice can be highly specific to a particular genetic background ., For example , >80% of hypomorphic MCM4 ( Mcm4Chaos3/Chaos3 ) females succumb to mammary adenocarcinomas when carried on the C3H genetic background on which it was isolated 4 ., In contrast when bred into a C57Bl/6J background females are highly prone to histiocytic sarcoma 7 ., Similarly , MCM2 deficient ( Mcm2IRES-CreERT2/ IRES-CreERT2 ) mice exhibit near to 100% penetrance of T-lymphocytic leukemia ( TLL ) within 5 months on the 129Sv background on which they were constructed 5 but a broader spectrum of tumor types with more variable latencies in a mixed 129Sv/BALB/c genetic background 8 ., The genetic lesions that occur in tumors arising in MCM deficient or hypomorphic mice are largely copy number alterations ( CNAs ) with a preponderance of deletions averaging 400–500 kbp in size 9–10 ., Several recurrent deletion sites are found in TLLs from MCM2 deficient mice where , on the 129Sv genetic background , the Tcf3 gene on Chr10 is affected in all tumors examined 9 ., Tcf3 is a transcriptional regulator that plays a central role in T-cell differentiation and prior studies have shown that loss of Tcf3 is sufficient to drive TLL in mice 11 ., Importantly , the Tcf3 gene lies within a region of Chr10 that shows preferential loss of origin function in MEFs from MCM2 deficient mice 12 ., Replication licensing may become limiting due to genetic deletion of Mcm2-7 , which occurs in many tumors 12 , oncogenic stress 1 and during stem cell aging 13 ., It is not known if under these conditions genome instability is elevated generally across the genome or whether specific locations are at increased risk; although loss of MCM expression in aging hematopoietic stem cells ( HSCs ) has been correlated with nucleolar associated DNA damage foci 13 ., In the present study we use micronuclear DNA sequences to show that the region of Chr10 carrying the Tcf3 gene is also the site of elevated ongoing genome instability even prior to tumor initiation in MCM2 deficient mice on the 129Sv genetic background ., The increased instability at this site predicts the high rate of genetic lesions affecting the Tcf3 gene and consequent high rate of TLL in these mice ., In addition to loss of Tcf3 , MCM2 deficient mice exhibit genetic lesions within the 45S ribosomal RNA gene repeats clusters on Chrs 12 , 16 , 18 and 19 consistent with the observed nucleolar associated DNA damage foci in aging HSCs ., The present study demonstrates that the consequences of reduced MCM expression on local genome instability is reflected in micronuclear DNA sequences allowing prediction of specific genetic lesions in the etiology of cancer and during aging ., To map sites of ongoing genomic instability across the genome we take advantage of the fact that in the hematopoietic system of mammals DNA remnants resulting from genetic damage events during differentiation of hematopoietic stem cells ( HSCs ) to erythrocytes are retained in the cells as micronuclei following enucleation 14 ., To isolate micronuclear DNA , cells from whole blood were first pelleted and washed to remove serum and serum DNAs ., Cells were then fractionated into lymphocyte and red blood cell ( RBC ) plus granulocyte fractions using ficoll-paque ., The RBC/granulocyte pellet was then re-suspended , RBCs were lysed using standard RBC lysis protocols and granulocytes were pelleted ., DNAs were recovered from three fractions: lymphocyte ( WBC ) , granulocyte ( GRN ) , and the supernatant from the lysed RBCs ( containing the RBC micronuclei , MN ) ., Tagged sequencing libraries were prepared from each sample to allow multiplexed high throughput sequencing such that WBC , GRN , and MN from the same animal are run together on the same sequencing lane on an Illumina HiSeq 2500 ., The resulting sequences were mapped and wiggle files were generated ., The sequence coverage , genome wide , from the MN fraction of a wild type ( wt ) 129Sv mouse is shown in Fig 1a ( for comparison , sequence coverage from the WBC and GRN genomic DNAs are shown in S1 Fig panels a and b respectively ) ., Sequence tag density in the WBC and GRN fractions of wt mice are largely uniform both between and across individual chromosomes ., In contrast , in the MN fraction , although the minimal sequence tag density is similar between chromosomes , there is a significant deviation from the average sequence tag density as a function of position across each chromosome ., This , in part , reflects the frequency with which different portions of each chromosome are present in micronuclei ., Similar patterns are seen in the MN fractions of two different wt mice where the correlation between experimental repeats is 0 . 988 ., Several parameters are defined to describe the observed changes quantitatively ( Fig 1b and 1c ) ., First , a value α is defined as a base line measure of the representation of each whole chromosome ., This parameter is specific to each chromosome and is expected to reflect both the whole chromosome loss rate for that chromosome in MN plus any whole genomic DNA contamination from nucleated cells ., The value of α is estimated from the minimum of the sequence tag coverage plot across the chromosome ., A value β is defined to describe the observation that sequence tag density increases over all chromosomes as a function of distance from the centromere ., This observation is consistent with double strand DNA breaks ( DSBs ) resulting in exclusion from nuclei of acentromeric DNA fragments from the breakpoint through the distal end of the chromosome 14 ., A third phenomenon , which is the converse of β , is referred to as ρ and describes a small increase in sequence tag density as a function of distance relative to the centromere distal telomere of each chromosome and which is observed primarily on the longest chromosomes ( Chr1 and Chr2 ) ., The mechanism leading to the observed ρ effect is unclear but may reflect mis-segregation or breakage of dicentric chromosomes resulting from errors during DNA repair/translocation 14–15 ., The contributions of α , β and ρ to MN sequence tag density , as a function of position on the chromosome , are shown schematically in Fig 1b ., In addition to chromosome wide changes , there are discrete locations on most chromosomes that show local increases in sequence tag density that are greater than those that would be expected based on the effects of α , β and ρ ., These locations are described by γ which is defined as localized changes in sequence tag density that are cumulative distal to the centromere even through the rate of change returns to levels predicted by β and ρ ( Fig 1c ) ., In principle , increased γ values are expected to reflect localized regions ( hot spots ) of the genome where chromosome breaks occur at increased frequency ., Functions describing β and ρ for wt animals were established using Chr7 and Chr11 since , by inspection , there are few localized increases in sequence tag density on these chromosomes ., Both β and ρ are non-linear where the best fit for each is a quadratic function ( S5 Fig panel, a ) ., It is unlikely that the non-linearity results from a bias in the locations of the initial DSB sites and consistent with this interpretation DSBs that have been repaired by translocation do not exhibit a bias that would lead to β or ρ distributions in micronuclei 15 ., One possible explanation is that events occurring during anaphase distort the representation of acentromeric chromosomal fragments in the micronuclear fraction ., For example , longer stretches of sister chromatid pairing within the acentromeric region resulting from DSBs more proximal to the centromere may promote non-disjunction and retention of the acentromeric fragment within a nucleus ., This could lead to preferential representation of acentromeric chromosome fragments resulting from breaks nearer to the centromere distal ends of the chromosomes consistent with the observed non-linear increase described by β ., To identify localized regions of the genome exhibiting elevated instability , γ values were estimated by determining the rate of the sequence tag coverage change within smoothed and normalized 20 kbp windows genome wide ., Although the overall contribution of β and ρ to differential representation of centromere distal and , to a lesser extent proximal ends of whole chromosomes is significant over entire chromosomes , over shorter 20 kb intervals these effects are minimal and the slope of the sequence tag density largely reflects the localized effect γ ( e . g . Fig 1d , γ track ) ., Similar to the case for sequence independent breaks , the magnitude of peaks identified by γ is biased towards the centromere distal ends of the chromosomes ., This result is expected if the same forces that act to skew the distribution of sequences represented in micronuclei following sequence non-specific breaks also act on chromosome fragments resulting from local hot spots ., The effect of normalizing γ peaks using β and ρ values is shown in Fig 1d , γ-normalized track ., γ plots show 294 peak locations across the genomes of wt mice ., Of these 129 occur in early replicating gene rich regions of the genome and 165 occur in gene poor late replicating , regions of the genome ., Prior studies have shown that agents that inhibit replication fork progression lead to chromosome breakage at specific locations across the genome referred to as common fragile sites ., Characteristics of these sites have been defined where breakage typically occurs at large , late replicating , transcriptionally active genes 16 ., To determine if Mic-Seq identifies common fragile sites , Mic-Seq was performed on hydroxyruea ( HU ) treated mice ., HU induces replication stress through inhibition of ribonucleotide reductase which leads to reduced nucleotide pools and increased replication fork stalling 17–18 ., To establish an informative dose of HU when administered in the drinking water a titration was performed where mice were assayed at both 1 week and 3 weeks of treatment with different HU doses for micronuclear frequency by FACS ( S2 Fig panel, a ) and at 3 weeks for effects on the levels of various cell types within the blood by CBCs ( S2 Fig panel, b ) ., The data shows that there is a narrow HU concentration window at which micronuclear frequency is elevated ( by ~10 fold ) during the interval between 1 and 3 weeks , but which has minimal effects on the frequency of various blood cell types ., The short half-life of HU 19 and intermittent dosing resulting from administration in the drinking water makes it likely that only a subset of cells , in various stages of S-phase , is transiently exposed to sufficiently high concentrations of HU to induce damage ., The observation that the frequency of micronuclei continues to increase for at least three weeks is consistent with this possibility but also suggests that , once formed , micronuclei are maintained stably in circulating RBCs ., Part of the efficacy of Mic-Seq analysis likely depends on the ability to accumulate DNA remnants over a period of time ., Mice treated with a dose of 2 mg/ml HU , which resulted in 1–2% micronucleated RBCs , were used for Mic-Seq ., Mic-Seq was performed on two mice treated with this dose and an additional untreated wt control ., Genome-wide sequence tag distributions are shown in Fig 2 for the MN fraction of both of the HU treated mice and is summarized for all three animals in Fig 3a ., The micronuclear fractions from HU treated mice show a more rapid increase in sequence tag density as a function of distance from the centromere relative to wt mice and the value estimated for β ( again using Chr7 and Chr11 ) is increased ~5 fold ( S5 Fig panel, a ) ., The increase in β is accompanied by a decrease in the value for ρ suggesting that more frequent chromosomal breaks suppress the mechanism that results in preferential retention of centromere proximal sequences ( S5 Fig panel, a ) ., HU also affects the distribution of MN sequences across whole chromosomes where Chrs 12 , 13 , 16 , 18 , 19 and X show an ~2 fold , increases in α values ( Fig 3b ) ., The 45S rRNA gene sequences , which are present on a subset of these chromosomes ( see below ) , also show a modest ( ~5–20% ) overrepresentation in the ratio of rDNA sequence tags to major satellite sequence tags in MN DNA relative to WBC or GRN DNAs from the same mice ( Fig 3c ) ., The sequence tag density of major satellite sequences is more strongly enriched ( ~ two-fold ) relative to minor satellite sequences in HU treated mice consistent with preferential induction of breaks in at least a subset of the major satellite sequences by HU ( Fig 3d ) ., Inspection of the micronuclear sequence data from HU treated mice reveals sharp increases in sequence tag coverage distal to discrete locations suggesting that specific sites are disproportionately affected by HU treatment ( 5 such locations are marked 1–5 in Figs 2 and 3a ) ., Extraction of γ peak values identifies these and additional localized increases in sequence tag densities that include 5 of the 8 molecularly characterized common fragile sites in the mouse ( e . g . site 2 is Wwox , S2 Fig ) and additional large , late replicating , transcriptionally active genes with properties of common fragile sites ( 16; Fig 4a–4c and S2 Fig ) ., In these cases , instability occurs within sub-domains of the genes ( Fig 4a and 4b and S2 Fig , panel c ) consistent with the fragile site core regions observed in prior studies 15 , 20 ., To examine the effect of insufficient DNA replication licensing on genome instability , Mic-Seq was performed on mini-chromosome maintenance ( MCM ) protein 2 deficient mice 5 ., MCM2 deficient mice on the 129Sv genetic background exhibit early onset T-lymphocytic leukemia , loss and dysfunction of stem cells , and genome instability evidenced by increased γH2AX in nucleated cells and increased micronuclei in reticulocytes/erythrocytes 5 ., The elevated frequency of micronuclei is confirmed in S3 Fig , panels 3a-3c , demonstrating that micronuclei are approximately 10 fold more frequent in RBCs of MCM2 deficient relative to wt animals ., For Mic-Seq , MCM2 deficient mice between 5–6 weeks of age , well before the onset of overt disease , were used ., MN sequence tag density across the genome of an MCM2 deficient 129Sv mouse is shown in Fig 5a ., Two biological repeats of the experiment were performed where the genome-wide correlation between experimental repeats was 0 . 975 ., The sequence coverage across all chromosomes is show for both experiments and in comparison to two wt mice in Fig 5 panel b ., Comparison of the sequence tag densities derived from micronuclei of wt and MCM2 deficient 129Sv mice shows that the additional breaks resulting from MCM2 deficiency suppress ρ values and modestly increase β values relative to wt cells ( S5 Fig panel, a ) ., However , the most pronounced differences are on the α values associated with chromosomes 6 , 12 , 18 and 19 , and to a lesser extent on chromosomes 15 and 16 , where these chromosomes are over represented by between ~1 . 8–12 fold relative to the average sequence tag density across the genome ( Fig 5c ) ., The entire mapped region of each of these chromosomes is affected ., Unlike the case for HU treatment , the ratio between minor and major satellite sequences is not affected ( S3 Fig panel g ) suggesting that breaks within the major satellite elements are not responsible ., In C57Bl6 mice , chromosomes 12 , 15 , 16 , 18 and 19 carry nucleoli encoding the 45S ribosomal rRNA gene repeats at centromere proximal positions ( NCBI http://www . ncbi . nlm . nih . gov/gene/19791 ) ., These are typically composed of 30–40 repeats of an approximately 45 kb repeating unit at each location ( additional low copy number rDNA repeats have also been mapped to Chr1 , Chr6 , and Chr9 in some strains; 21 ) ., One potential explanation for the over representation of the subset of chromosomes seen in micronuclei from MCM2 deficient mice is that rDNA repeats are hypersensitive to reduced replication origin licensing resulting in high rates of double strand breaks within these repeats ., Since rDNA repeats are adjacent to the centromeres in these chromosomes , a DSB within an rDNA repeat would render nearly all of the affected chromosome acentromeric ., To determine the locations of rDNA clusters in 129Sv mice , metaphase chromosomes from wt 129Sv MEFs were assayed by fluorescence in situ hybridization ( FISH ) plus spectral karyotyping ( SKY ) to localize a probe for the 45S ribosomal gene ( containing portions of the 18S , 5 . 8S and 28S ribosomal genes ) to specific chromosomes ., Results from these studies ( S3 Fig panels 3d-3e ) localize ribosomal gene repeats to Chrs 12 , 16 , 18 and 19 in most metaphase spreads from 129Sv mice ., In addition , a small subset ( 2% ) exhibit signal consistent with recombination events leading to the presence of rDNA repeats on Chr6 ( and in MCM2 deficient MEFS , Chr15 ) ., Although these results are consistent with a role for rDNA repeats in the increased representation of specific chromosomes within micronuclei , the relative FISH signal intensity between different chromosomes ( 19 = 12>18 = 16 , S3 Fig panel, f ) does not correlate with the representation of these chromosomes in MN of MCM2 deficient mice ( 19>>12 = 18>>16 , Fig 5c ) ., However , if the increased representation of the mapped regions of these chromosomes results from DSBs in the rDNA repeats , representation of the acentric regions is expected to be affected by β ., Following normalization for β , the representation of different chromosomes in MN is similar to that expected based on rDNA copy number as estimated from FISH ( i . e . 19 = 12>18 = 16 ) ., rDNA sequences are over represented by a factor of 2–3 in the MN fraction , relative to the GRN or WBC fractions , of MCM2 deficient but not wt mice ( Fig 5d ) ., Further , rDNA sequences are enriched by a factor of 2–5 fold relative to peri-centromeric ( major satellite , Fig 5e ) and centromeric ( minor satellite ) repeat sequences in MN of MCM2 deficient mice in comparison with the WBC or GRN fractions of the same mice or all fractions from wt animals ., These results support a large increase in the rate of DSBs in at least a subset of rDNA repeats of MCM2 deficient relative to wt mice ., Consistent with this interpretation , many of the additional γH2AX foci observed in MCM2 deficient , relative to wt , MEFs are located over nucleoli ( S4 Fig panels a-g ) ., Further , short nascent stand analysis shows that a subset of DNA replication origins within the 45S rRNA gene repeats are preferentially affected by MCM2 deficiency ( S4 Fig panels h-j ) ., To examine the effect of MCM2 deficiency on localized increases in genetic damage in regions other than the 45S rRNA gene repeats , γ values were compared between wt and MCM2 deficient mice genome-wide ( e . g . Chr10 , Fig 6a ) ., Most locations where peaks are present in the γ plot of MN from wt mice are also represented in MN from MCM2 deficient mice ( 214 common peaks across the autosomes ) but at increased values where the average increase was 2 . 4 fold ( Fig 6b , blue diamonds ) ., An additional 63 peaks are present only in MN from MCM2 deficient mice ( Fig 6b , green circles ) ., Prior studies 9 have identified locations of recurrent deletions in T-lymphocytic leukemias ( TLLs ) arising in MCM2 deficient mice ., One location that undergoes deletion in all TLLs of MCM2 deficient mice on the 129Sv genetic background is the Tcf3 gene on Chr10 and this location is highlighted in Fig 6a ., Tcf3 is required for T-cell differentiation and loss of Tcf3 results in TLL 11 ., This site is also a location at which MCM2 deficiency has a disproportionately strong effect in reducing origin usage as measured by short nascent strand analysis 12 ., Increased genome instability is detected by Mic Seq at this site and , of the early replicating regions of the genome , the region containing the Tcf3 gene shows the highest level of instability genome wide ( Fig 6c ) ., It is likely that instability at this site drives a high rate of loss of Tcf3 resulting in the near 100% penetrance of early onset TLLs in these mice ., Smaller local increases in γ values are also found at many of the additional recurrent deletion sites found in these tumors ( Fig 6c ) ., However , locations where MCM2 deficiency has the largest effects on representation of sequences in MN , and results in the greatest increases in γ peak values , occur in preferentially in late replicating gene poor regions of the genome that are not the sites of recurrent deletions in TLLs ., Chromosomal fragile sites are genomic locations that are hotspots for genome instability leading to translocations , amplifications and deletions ., Such locations were first defined cytogenetically following treatment of cells in culture with agents that impede DNA polymerase ( including aphidicolin , hydroxyurea , 5-azacytidine and bromodeoxyuridine ) and mapping breaks in banded metaphase chromosomes ., Numerous studies have characterized sites that are frequently affected in human and mouse cells and led to identification of a set of locations termed common fragile sites that are affected under conditions of chemically induced replication stress in a high proportion of individuals ., These sites have been extensively characterized and are significantly associated with the presence of large , transcriptionally active , and late replicating genes over 300 kbp in size 16 ., The frequency of chromosome breaks at these locations is dependent on cell type and the specific agent used to induce replication stress ., Here we have used the representation of different genomic regions in micronuclear DNA sequences to infer the frequency of chromosome breaks in erythroid cells in vivo ., Many micronuclear sequences result from the presence of double strand DNA breaks that lead to failure of acentromeric portions of chromosomes to segregate to the nucleus during mitosis ., Further we take advantage of the fact that micronuclei are retained in maturing RBCs following enucleation in mammals ., In contrast to defining fragile sites cytogenetically , harvesting micronuclei from RBCs allows recovery of tens of millions chromosomal remnants each of which defines a breakpoint that can be queried at nucleotide level resolution by high throughput sequencing ., Although there is a concern that erythroblasts ( just before enucleation ) may not reflect the DNA repair and checkpoint responses typical of other cells , application of the Mic-Seq method to define fragile sites in mice treated with hydryoxyurea shows that the method identifies 5 of the 8 molecularly characterized fragile sites previously defined in mouse lymphocytes 22 including those occurring at the Wwox and Immp2 genes ., Further sites detected in this study support that the majority of the most sensitive HU induced fragile sites occur within subdomains of the transcribed regions of large ( >300 kbp ) , transcriptionally active , genes similar to prior studies and consistent with the possibility of interference between the transcription and replication machinery under conditions of DNA polymerase inhibition 15–16 , 20 ., Similar to HU treatment , reduced replication licensing results in genome instability , increased DSBs , and chromosomal deletions and rearrangements 2–5 , 9 ., The mechanism resulting in this damage is likely to differ from that mediating HU induced damage; however , it has not been previously determined whether similar or different locations across the genome are preferentially affected ., In this study we demonstrate that MCM2 deficient mice exhibit a distinct sequence tag distribution profile in Mic-Seq relative to HU treated mice ., HU induced fragile sites are not preferentially sensitive to MCM2 deficiency ( e . g . Fig 4 , panels a and b ) and there is little overlap with early replicating fragile sites induced by higher concentrations of HU 23 ., Unlike HU treated mice , the locations at which MCM2 deficient mice exhibit localized damage by Mic-Seq analysis are largely locations that are already sensitive in wt cells but become more prone to breakage in MCM2 deficient mice ., Differences in the genome instability profiles are likely to reflect differences in the mechanisms by which HU and MCM2 deficiency affect genome stability ., Prior studies have shown that , unlike HU or aphidicolin , reducing MCM levels does not affect the rate of DNA polymerization since similar tract lengths of CldU or IdU incorporation are found between wt and MCM deficient cells by DNA fiber analysis 2–3 , 8 ., However , DNA fiber analysis also shows that MCM deficient cells are unable to initiate DNA synthesis from dormant origins under conditions ( HU treatment ) that lead to replication fork stalling 2–3 , 8 consistent with a reduction in the frequency of licensed origins ., Short nascent strand analysis has shown that origin usage does not decline uniformly across the genome , but rather specific locations lose function preferentially , in MCM2 deficient MEFs 12 ., These locations tend to occur in gene rich , early replicating , regions of the genome although they are not exclusive to transcribed regions of active genes and include origins within inactive genes and intergenic regions 12 ., In contrast , Mic-Seq localizes sites that are preferentially prone to breakage to subsets of both early and late replicating regions ., This result suggests that factors in addition to the degree of reduction in replication licensing affect the rate of chromosome breaks under conditions of MCM2-7 deficiency ., However , even in late replicating regions of the genome these factors are not the same as those determining fragile site locations following HU treatment since the locations that become sensitized to breakage by MCM2 deficiency are not known common fragile sites , or large genes generally , and many contain few or no transcribed regions ., Even within early replicating regions of the genome chromosomal breaks detected by Mic-Seq show only a modest correlation with locations where SNS analysis demonstrates that origin usage is most affected ., Nonetheless , it is important that many of the locations showing recurrent deletions in the TLLs that arise in MCM2 deficient mice 9 are sites at which there is both a preferential loss of origin function 12 and an increase in DSBs detected by Mic-Seq ., In particular , loss of Tcf3 is sufficient to drive TLL formation 11 and the region carrying this gene shows a strong differential signal between wt and MCM2 deficient mice in both SNS and Mic-Seq analyses ., These differences are apparent in samples taken well before tumorigenesis is observed , likely before tumors are initiated , and support that an increased DSB rate detected by Mic-Seq can in some instances predict a high probability of tumor occurrence ., The strongest signals observed by Mic-Seq in MCM2 deficient mice are associated with chromosomes carrying nucleoli and implicate a high rate of DSBs within rDNA clusters as the location of much of the DNA damage , and the source of a large proportion of the additional micronuclei , found in MCM2 deficient mice ., This observation confirms and extends prior studies showing that , as mice age , MCM levels are suppressed in HSCs and , coincident with the loss of MCM expression , nucleolar associated γ-H2AX foci accumulate 13 ., The presence of unidirectional replication fork barriers reviewed in 24 may sensitize rRNA gene repeats to loss of licensed replication origins ., These results demonstrate that experimental reduction of MCM proteins in young asymptomatic mice is sufficient to cause a profile of genome instability that predicts at least a subset of chromosomal locations where genetic damage is found in both cancers and during aging ., The observation that , unlike HU , many locations where MCM2 deficiency causes instability can already be detected in young wt mice suggests that even under normal conditions the distribution of licensed DNA replication origins contributes to base line levels of chromosomal instability ., Although replication stress is widely recognized as a potent cause of genomic instability 1 , 25 , the present study emphasizes that different mechanisms leading to replication stress have very different consequences for instability at various locations across the genome and can result in very different phenotypic outcomes ., Animal husbandry programs and protocol reviews are in compliance with NIH , USDA , and New York State Standards ., Mice were maintained in facilities covered under NIH assurance #A-3143-01 , certified by New York State for the use of living animals , and the USDA APIHS registration as research facility #21–124 ., The studies were approved by the Roswell Park Cancer Institute Animal Care and Use Committee under Protocols 817M and 876M ., Five to six week old wild type 129Sv and Mcm2 IRES-CreERT2/IRES-CreERT2 ( MCM2 deficient ) mice were used in studies addressing the effects of MCM2 deficiency ., For studies addressing the effects of hydroxyurea ( HU ) , 3 month old wild type 129Sv mice were administered HU continuously in the drinking water at the concentrations indicated in the text ., Blood samples were taken by retro-orbital bleed or cardiac puncture ., For flow cytometric analysis of micronuclei 26 blood samples were fixed in methanol on the day of sample collection and processed for flow cytometry using the Litron MicroFlow plus kit for mouse blood as per the manufacturer’s instructions ( Cat . No . 552730 , BD Biosciences ) ., Combined SKY/FISH was performed on wt or MCM2 deficient mouse embryonic fibroblast ( passage 3 ) by the Roswell Park Cancer Institute SKY/FISH core facility ., The rDNA probe for FISH analysis was prepared using Nick Translation Reagent Kit 07J00-001 ( Abbott Molecular Inc . ) Green-dUTP 02N32-050 ( Abbott Molecular Inc . ) to fluorescently label a 7109 bp EcoRI fragment from human genomic ribosomal gene DNA containing a portion of the 18S ribosomal RNA gene , the intergenic spacer , the 5 . 8S ribosomal RNA gene and a portion of the 28S ribosomal RNA gene ., Between 400–500 μl of whole blood was washed with 10 ml of phosphate buffered saline ( PBS ) 3 times and re-suspended in 3 ml PBS ., The sample was then layered over 2 ml of Lymphocyte Separation Medium ( density = 1 . 077–1 . 080 g/ml; Mediatech Inc . ) in a 15 ml centrifuge tube and spun for 15 min at 800 RPM ., Lymphocytes ( WBCs ) at the PBS-media interphases were collected and placed in a 15 ml tube and washed 3 times with 10 ml PBS prior to pelleting for DNA isolation ., Separation media was removed from the red blood cell ( RBC ) /granulocyte ( GRN ) pellet in the original tube and cells were washed 3 times with 10 ml PBS and pelleted ., The cell pellet was then resuspended in 4 ml of RLF lyse buffer and incubated at room temperature for 5 minutes prior to spinning at 800 RPM for 5 minutes ., The supernatant was collected as the RBC micronuclear

Introduction, Results, Discussion, Methods, Data Access

Insufficient licensing of DNA replication origins has been shown to result in genome instability , stem cell deficiency , and cancers ., However , it is unclear whether the DNA damage resulting from deficient replication licensing occurs generally or if specific sites are preferentially affected ., To map locations of ongoing DNA damage in vivo , the DNAs present in red blood cell micronuclei were sequenced ., Many micronuclei are the product of DNA breaks that leave acentromeric remnants that failed to segregate during mitosis and should reflect the locations of breaks ., To validate the approach we show that micronuclear sequences identify known common fragile sites under conditions that induce breaks at these locations ( hydroxyurea ) ., In MCM2 deficient mice a different set of preferred breakage sites is identified that includes the tumor suppressor gene Tcf3 , which is known to contribute to T-lymphocytic leukemias that arise in these mice , and the 45S rRNA gene repeats .

Many RBC micronuclei result from double strand DNA breaks that give rise to acentromeric chromosomal fragments that fail to incorporate into nuclei during mitosis and consequently remain in the cell following enucleation ., Here , RBC micronuclear DNA is sequenced ( Mic-Seq ) to define the locations of breaks genome-wide and this assay is used to study ongoing genome instability resulting from insufficient DNA replication origin licensing ., Using a mouse model , we show that there is increased instability at discrete sites across the genome , which include genes that are recurrently deleted in the T-lymphocytic leukemias that eventually arise in these mice ., Mic-Seq may provide an effective means of predicting locations that are susceptible to genetic damage and these predictions may have prognostic value .

genetic networks, micronuclei, microbiology, sequence tagged site analysis, dna replication, network analysis, protein structure, mammalian genomics, dna, research and analysis methods, sequence analysis, computer and information sciences, chromosome biology, bioinformatics, proteins, repeated sequences, protein structure networks, molecular biology, animal genomics, biochemistry, macromolecular structure analysis, cell biology, nucleic acids, database and informatics methods, gene identification and analysis, genetics, biology and life sciences, protozoology, genomics, chromosomes

journal.pcbi.1003748

Large Scale Characterization of the LC13 TCR and HLA-B8 Structural Landscape in Reaction to 172 Altered Peptide Ligands: A Molecular Dynamics Simulation Study

Recognition of immunogenic peptides presented by Major Histocompatibility Complex ( MHC ) molecules to the T-cell receptor ( TCR ) of T-cells is a key event in the adaptive immune response ., In order to achieve this recognition process , a peptide in the MHC class I pathway will go through several processing steps 1 ., First , a protein is degraded into peptide fragments by the proteosome ., Second , the peptide enters the endoplasmic reticulum ( ER ) via the “transporter associated with antigen processing” ( TAP ) or alternative pathways such as Sec61 2 ., Third , a potential epitope must bind to the MHC class I molecule ., Finally , this peptide/MHC ( pMHC ) complex is presented at the cell surface where its recognition by the complementary determining regions ( CDRs ) of TCRs can take place ., Predicting whether a peptide will undergo the initial steps ( one and two ) outlined above has been shown to have only a minor impact on the quality of T cell epitope prediction ., This is probably due to the inability to accurately model these processes 3 , 4 ., In contrast , the prediction of the binding between peptide and MHC is well understood and frequently utilized for the prediction of potential T cell epitopes ., Pan-specific peptide/MHC binding affinity prediction methods have reached coverage of almost all MHC class I 5 and class II 6 alleles ., However , while the affinity prediction accuracy is high and binding affinity between peptide and MHC is a commonly used indicator for peptide immunogenicity 3 it is known that binding between peptide and MHC is necessary but not sufficient for T cell activation 7–13 ., It is an obligatory prerequisite i . e . If a peptide does not bind or only binds very weakly to MHC , the necessary density of cell surface pMHC cannot be reached and T cell activation cannot take place ., However , a peptide binding strongly to an MHC is no guarantee of T cell activation ., Thus it is necessary to predict peptide immunogenicity , but this has proved far more challenging than prediction of the peptide/MHC binding affinity ., This is mainly due to a limited understanding of which properties determine an MHC-binding peptide as immunogenic in contrast to peptides binding to the same MHC but being non-immunogenic 12 ., The question “Which parameters are the driving force behind T cell activation ? ” is still a matter of frequent discussion and not understood in detail 14 ., Suggested determinants range from binding affinity , association and dissociation rates , and half-life of interaction 11 to structural adjustments in the TCR/pMHC interface 10 , 15 , amino acid preferences 12 , changes in heat capacity 16 , similarity in biochemical properties 17 , hydrophobicity , molecular weight , and structural patterns in the peptide 9 ., There are two major hypotheses for T cell activation 11 , 14: ( 1 ) The affinity model i . e . the number of TCRs binding to pMHCs is the most important factor and ( 2 ) the half-life model i . e . the TCRs must bind to pMHC with a certain binding affinity and duration ., Another proposal groups models for TCR triggering into aggregation models , conformational change models , and segregation models 18 ., Despite all the advances over the last few years there is still much to learn about MHC class I restricted immune responses 19 ., Methods for the prediction of peptide immunogenicity as opposed to peptide/MHC binding affinity , are rare 3 and show limited accuracy ., Recently the method POPISK 9 has pioneered the field of immunogenicity predictors ., However , an independent evaluation yielded almost random results for this method ( AUC 0 . 52 and 0 . 49 ) 12 ., Although it seems that some amino acids , especially large and aromatics ( e . g . W , F , I ) , are likely to be associated with peptide immunogenicity , the predictive power of their presence is quite limited 12 ., Therefore , since peptide immunogenicity is hard to explain from the peptide sequence alone research groups have investigated the spatial dynamics of ( TCR ) pMHC complexes computationally ., Many immuno-informatics studies have used MD simulations to investigate the spatial dynamics of different systems ., Experiments have included the use of the same MHC with different peptides 10 , 20–31 , different MHCs with the same peptide 20 , 25 , 32 , 33 , the same peptide/MHC complex with different TCRs 24 , 27 , simulations including trans-membrane regions 22 , 34 , peptide free simulations 29 , 35 , 36 , steered simulations 27 , 30 , and single simulations 37 , 38 ., In most of these cases the real immunological outcome is known ., Subsequently the differences between runs have been compared on the basis of typical MD simulation descriptors ., For example Reboul et al . 33 performed 100 ns simulations of HLA-B*35:01-LPEP , HLA-B*35:08-LPEP , and SB27-HLA-B*35:08-LPEP ., The SB27 interaction with HLA-B*35:08-LPEP induces a cytotoxic T-cell response while the interaction with HLA-B*3501-LPEP does not ., In their simulations the authors find an increased flexibility of the peptide bound to HLA-B*35:01 and propose that difference to impede productive interaction with SB27 and therefore hamper cytotoxic T-cell response ., In another study , Narzi et al . 20 used 400 ns MD simulations to investigate the ankylosing spondylitis-associated HLA-B*27:05 as well as the non-ankylosing spondylitis-associated HLA-B*27:09 with one viral and three self peptides ., They found an increased entropy for the viral peptide presented by the disease associated MHC allele ., For the same allele they find enhanced flexibility of the α1-helix which they hypothesize to be important for receptor binding ., In complementary work Kumar et al . 25 performed 120 ns simulations of the multiple sclerosis predisposing allele DRB1*15:01 and the protective allele DRB1*16:01 ., Both alleles were simulated in combination with a myelin basic protein peptide as well as Epstein Barr Virus derived peptide ., The predisposing allele formed a stable complex with both peptides ., In contrast the protective allele did not form a stable complex with the virus peptide ., In another investigation Stavrakoudis 37 simulated the same structure as used in this study ., He performed a single 20 ns simulation of the wild-type peptide and found two conformational clusters in the peptide structure as well as that the TCRpMHC interface becomes increasingly solvated over simulation time ., In a previous study we have used MD simulations to support experimental peptide/MHC binding affinity and T cell activation data measured by collaboration partners ., In a mugwort pollen allergen model we compared core-identical 12-mer and 18-mer peptides bound by HLA-DR1*01:01 with the outcome of 20 ns MD simulations 21 ., In a second study we compared the experimental results of altered versions of the 12-mer in complex with HLA-DR1*01:01 and HLA-DR1*04:01 with the outcome of 30 ns simulations 32 ., In all of the studies mentioned above only a small number of simulations were run ., This approach tends to be suboptimal since two MD simulations will always differ in some aspects if the simulation time is of finite length ., Even two identically parameterized simulations ( using the same initial seed ) might produce different trajectories due to parallelisation and floating point imprecision ., On this basis and the fact that a TCRpMHC system consists of roughly 8000 heavy atoms one might always be able to describe some differences between two individual simulations ., This problem is made worse as the differences between simulations may be real but unrelated to the immunogenicity of the peptide ., For example if an MHC with a very immunogenic peptide and an almost identical non-immunogenic peptide whose position seven is point mutated with a smaller amino acid yield differences , one might not be able to distinguish whether these differences result from a size change at position seven or from different peptide immunogenicity ., If one wants to address this issue then a frequent challenge is the choice of appropriate experimental data ., This is problematic because some experimental findings may be false positives 39 , 40 , hard to reproduce , and/or not comparable with other experimental results 41 ., These problems might be caused by unknown marginal differences in the experimental conditions , unintended human influence , or just different consumables used ., On this basis the first aim for a systematic and large scale characterization of TCRpMHC interaction is to find an appropriate test set with experimental immunogenicity data ., For our study we selected the data from Kjer-Nielsen et al . 42 because all data was ( 1 ) determined using the same technique , ( 2 ) the same conditions , ( 3 ) published by the same group , ( 4 ) in the same manuscript , and ( 5 ) the data set contains a sufficient number ( 172 ) of systematic experimental immunogenicity values ., ( 6 ) In addition a crystal structure of exactly this complex was determined by the same group ( Protein Data Bank ( PDB ) 43 accession code 1mi5 42 ) ., Kjer-Nielsen et al . performed a fine specificity analysis of LC13 cytotoxic T cell ( CTL ) reactivity to all possible single substitution altered peptide ligands ( APLs ) of the Epstein Barr Virus ( EBV ) peptide FLRGRAYGL bound by HLA-B*08:01 ., The employed assay of Kjer-Nielsen et al . is described in more detail in 44 ., This yields a total of 172 experimental values ( 19 amino acid substitutions in 9 peptide positions and the wild-type ) ., To determine which and if any structural and dynamical factors are actually involved in peptide immunogenicity we performed 172 systematic 100 ns MD simulations of the TCRpMHC system described above ., This study is far larger than any other previous ( TCR ) pMHC MD study ., Our study assumes that if a structural or dynamical factor is involved in determining peptide immunogenicity it would be conserved across the peptides in our set ., It is possible that different factors play different roles for every peptide ., Our large scale test finds very little evidence of conserved structural and dynamical differences which could be related to peptide immunogenicity ., We used the above described crystal structure of the FLRGRAYGL peptide bound by HLA-B*08:01 and presented to the LC13 TCR ( PDB accession code 1mi5 ) as our basis ., We modelled all possible 172 single point APLs ., For each of these altered peptide ligands experimental immunogenicity data exists 42 ., For this purpose we used the software SCWRL 45 via the PeptX framework 46 to replace the side-chains ., We have previously shown that SCWRL is the most appropriate software in the context of peptide/MHC interactions 47 , 48 ., The peptides backbone structure is relatively conserved within the same MHC allele 47 and small expected changes induced by single side-chain substitutions are accommodated by the subsequent energy minimization ., The α1–3 regions of the MHC , the β2-microglobulin , as well as the variable and constant regions of the TCR were included in the models yielding 172 TCRpMHC complexes each consisting of 827 residues ., One such model complex is shown in Figure 1 ., All MD simulations were performed using GROMACS 4 49 and the GROMOS96 53a6 force field 50 ., Each of our 172 modeled structures was immersed into a separate dodecahedronic simulation box of 3410 nm3 volume which was filled with ∼107 , 750 explicit SPC water molecules allowing for a minimum distance of 1 . 5 nm between box boundary and protein ., Na+ and Cl− ions were also added to achieve a neutral charge and a salt concentration of 0 . 15 mol/liter ., Each of the systems was energetically minimized using the steepest descent method and then warmed up to 310 K . Finally , we conducted MD simulations of 100 ns on each of the systems using the ARCUS cluster of the Oxford Advanced Research Computing ( ARC ) facility ., This yields a total simulation time of 17 . 2 µs for our 827 residue systems ., Here we consider the first 10 ns of each simulation to be the initial relaxation time of the system ., All analysis is based on the last 90 ns ., Manual pre-inspection of the trajectories was carried out using the vmdICE plugin 51 of VMD 52 ., All graphical 3D representations were rendered in VMD ., The analysis of the hydrogen bonds , solvent accessible surface area ( SASA ) , and root mean square fluctuations ( RMSF ) were carried out using the GROMACS functions g_hbond , g_sas ( implementing 53 ) , and , g_rmsf respectively ., The “percent present” value of hydrogen bonds is a normalized frequency score which is zero if no hydrogen bond is present in any timeframe for this residue ., It is one if one hydrogen bond is present in all frames ., This score can exceed one if a residue mediates more than one hydrogen bond , as for example occurs for the anchor residues of the peptide ., The peptide/MHC binding affinities were calculated using the ligand/protein rescoring function XSCORE 54 which has been shown to be the most appropriate for structural peptide/MHC binding predictions 48 ., The binding affinity between TCR and pMHC was calculated using two protein/protein rescoring functions IRAD 55 and ZRANK 56 ., The relative orientation between the variable domains of the TCR , Vα and Vβ , was measured using a TCR-adapted version of the ABangle 57 methodology ( shown in Figure S1 ) ., Here , the orientation is described by five angles ( BA , AC1 , BC1 , AC2 and BC2 ) and a distance ( DC ) ., The length , DC , describes the distance between consistent points on the interface of the two domains ., The angle BA describes a torsion angle between the variable domains of the α and β-chains ., AC1 and BC1 are tilting-like angles of one domain towards the other ., AC2 and BC2 describe twisting-like angles of one domain with respect to the other ., The distances in the TCRpMHC interfaces were measured using the gro2mat package 58 ., Peptide immunogenicity is a continuous variable ., However , to compare the characteristic features of more immunogenic simulations to those of less immunogenic simulations we have introduced a discrete split between these groups ., Based on the experimental data of Kjer-Nielsen et al . 42 , the 51 peptides which never induce 50% lysis make up the less immunogenic group ( groupL ) ., The 51 most immunogenic peptides were designated as the more immunogenic group ( groupM ) ., Each member of groupM induces 50% lysis using a peptide concentration of 10−6 . 94 M or less 42 ., All results shown below are based on this split ., This is a subset consisting of the 102 most extreme cases of our 172 simulations ., The total variation distance ( tvd ) was used to quantify how strongly probability distributions of the above described parameters differ between our groupL and groupM sets ., The tvd is defined as:Where f1 ( x ) is the first distribution normalized and f2 ( x ) the second distribution normalized ., Thus tvd will range between 0 and 1 ., A value of 0 represents perfect overlap of the distributions while a value of 1 represents no overlap ., Since the tvd would yield a high value for identical means in combination with severely different variances we additionally calculate a normalized distance between the means of the distributions ., This value is referred to as d/r and defined as:Where and are mean value of the two distributions and the denominator is the range of the combined distributions excluding the lowest and highest 2 . 5% ., To further determine what values of tvd and d/r are relevant we performed permutation tests ., We compared the tvd and d/r of the groups under investigation against the distribution of 2000 random assignments of the simulation trajectories to the groups ., This approach is illustrated in Figure S2 ., If at least 90% of the permutations exhibit a smaller difference than the groups under investigation we refer to a slight difference ., If at least 95% we refer to a difference , if at least 99% we refer to a strong difference ., The binding affinities in the TCRpMHC interface are thought to be related to peptide immunogenicty ., We measured the binding affinity between peptide and MHC using XSCORE 54 on the basis of equally distributed frames extracted from the trajectories ., Although the mean binding affinity of groupM is lower than that of groupL this difference is not large ( Figure 2A ) ., We also calculated the binding affinity between pMHC and TCR using both IRAD 55 and ZRANK 56 , here the binding affinity distributions are also highly similar ( Figure 2B and 2C ) and no difference was found ., Analyzing the overall binding affinity did not yield differences between groupL and groupM ., A major contributor to the binding affinity and commonly used descriptor for MD simulations are the hydrogen bond ( H-bond ) footprints ., In our case: In which residues are H-bonds occurring most frequently during simulation and is there a difference between groupM and groupL ?, The footprint of the first frames ( before MD simulation , Figure S3 ) differs significantly from the MD footprint ( Figure 3 ) ., This highlights the importance of the dynamics of a system in contrast to static structures ., The H-bond footprint of the peptide to the MHC over simulation time recovers a key feature of the experimentally known binding profile ., The anchor residues for HLA-B*08 are known to be peptide positions three , five , and nine 59 ., In all three positions the number of H-bonds is higher in comparison to all other positions ( Figure 3A ) ., Additionally the number of H-bonds between the peptide and the TCR is increased around peptide position seven ( Figure 3B ) which has been described as the main TCR interaction site 42 ., In this plot it also seems that the number of H-bonds in peptide position four is far higher in groupL ., However , this can be explained by the fact that Gly is the wildtype residue for peptide position four and G4A and G4P are the two most immunogenic APLs ., These three residues have no ability to form side-chain H-bonds with the TCR ., The H-bond footprints between the TCR and the MHC during the simulations revealed a preference for H-bonds to occur mainly in the CDR regions ., It is experimentally known that these regions are the main interaction sites of the TCR for MHC binding 60 ., However , there is no sign that more immunogenic peptides induce a different H-bond footprint in the CDRs than less immunogenic ones ( Figure 3C ) ., A different behaviour is observed if the H-bond footprints between the MHC helices and the TCR are investigated ., While groupM has an increased number of H-bonds in MHC helix 1 , groupL has an increased number of H-bonds in helix 2 ( Figure 3D ) ., No mutations were introduced in the MHC helices , so the changed H-bond footprints in the MHC helices are all induced by the peptides which are adjacent to the helices but do not directly participate in the H-bonds between the helices and the TCR ., In addition to H-bonds , the size of the buried interface area of a binding site plays an important role in determining the mode of interaction and the binding affinity ., Therefore we calculated the SASA over simulation time for the CDRs , peptide , and MHC helices ., Whilst each CDR exhibits its specific distribution of SASA values there is no relevant difference between groupM and groupL and the mean values of the two groups are almost perfectly identical ( Figure 4 A–F ) ., Likewise , the distributions of the SASA values for the peptide and the MHC-helices are highly similar and no relevant difference between groupM and groupL could be found ( Figure 4G–I ) ., Another feature thought to play an important role in the immunogenicity of a peptide is the flexibility of the involved interface residues ., Many previous studies have hypothesized that an increased or decreased flexibility of certain components of the TCRpMHC interface is the reason for being more or less immunogenic ( see discussion ) ., Therefore we calculated the RMSF of the CDRs , peptide and MHC helices over simulation time ., While the RMSF of the CDRs is almost identical between groupM and groupL ( Figure, 5 ) the shapes of the RMSF curves recover a key feature of the interaction landscape ., The middle part of each CDR loop is exhibiting the highest amount of flexibility which is in agreement with the notion that the CDRs are highly polymorphic and dynamic pMHC binding probes ., The RMSF values of the peptide and the MHC helices are also highly similar between groupM and groupL ., We could not find an increased or decreased RMSF for either group ., Furthermore the peptide shows a similar amount of flexibility in all of its residues ( Figure 5G ) while the helices are generally more flexible at their N- and C-terminal ends compared to their middle part ( Figure 5HI ) ., Taken together these indicate that flexibility in the TCRpMHC interface is unlikely to be an important player in peptide immunogenicity ., Experimental crystallographic data have shown a wide variety in the binding mode and angle between the MHC and TCR ., Furthermore in-silico calculations have revealed that a state-of-the-art forcefield can reproduce these orientations 61 ., Therefore we investigated whether our groupM and groupL differ in their binding mode by calculating 16 distances in the TCRpMHC interface ( Figure, 6 ) for equally distributed individual frames of the trajectories ., The 16 distance distributions show two properties which are common to all simulations ( 1 ) no TCR detached or tilted away from any pMHC , ( 2 ) no major binding mode rearrangement took place within the interface itself ., In terms of differences between groupM and groupL we found that the distances between the central kink of MHC helix 2 are closer to CDR1a and CDR2a for more immunogenic peptides ( Figure 6D , F ) ., The distance between the middle of helix 1 and helix 2 is also decreased for groupM ( Figure 6H ) ., Only the distance between the N-terminal peptide end and CDR3b is decreased for less immunogenic peptides ( Figure 6M ) while the distance between the C-terminal peptide end and CDR3a as well as CDR3b is decreased for more immunogenic peptides ( Figure 6N , O ) ., The relative orientation between the two chains of a TCR is important for their binding mode , specificity , and affinity ., Therefore we investigated the relative orientation of the two TCR chains during simulation using the ABangle methodology 57 ., In this way we found a slight difference in the BA torsion angle as well as a difference in the BC2 twist angle and the DC that characterises the distance between the two variable domains ( Figure 7 ) ., In all three cases the mean value was smaller for groupM ., These differences between the groups represent only small physical changes in orientation of the TCR chains ., However , the differences that do arise suggest that groupM simulations have slightly more “open” binding site conformations best characterised by the larger ( more negative ) BA torsion angles ., Attempts to fully understand the interaction process between ( TCRpMHC ) have used MD studies ( see introduction ) ., However , all of these studies compare only a small number of simulations ., For example they compare the behaviour of one MHC with two different peptides ., Hence it is hard to determine if differences that are found actually relate to immunogenicity or not ., To address this challenge we present , to our knowledge , the largest systematic dataset of simulations of the TCRpMHC interface ., We present a dataset of 172 TCRpMHC simulations each of 100 ns ., The data shown in the results section consists of a subset of the 51 most immunogenic peptides compared to the 51 least immunogenic peptides ., As we are looking for a systematic difference between more and less immunogenic peptides we showed these sets rather than all 172 , as this should make such a difference easier to spot ., The results of these 102 ( 51vs51 ) simulations and the full set of 172 ( 82vs90 ) simulations are highly similar ( compare Figure 5 and Figure S4 ) ., Our dataset recovers several key features of the known TCRpMHC interaction landscape 60 ., We show that the number of H-bonds between the experimentally known anchor amino acids of the peptide and the MHC is significantly higher than in the other peptide residues ( Figure 3 ) ., Related , the number of H-bonds between the immunological hotspot of peptide position seven 42 and the TCR is significantly higher than for other peptide positions ., Also the H-bonds between TCR and MHC are almost exclusively formed by the CDR regions of the TCR ., Furthermore , the flexibility footprint of the CDRs ( Figure 5 ) is in agreement with the notion that these hypervariable and flexible regions are able to scan and complement the surface of pMHCs while the framework regions around them are more rigid ., No major structural defolding of TCR or MHC parts took place which is in agreement with known experimental TCRpMHC structures which have an overall conserved secondary and tertiary structure 60 ., Taken together the recovered key features support the view that current state of the art MD simulations are capable of investigating the relevant dynamics of a TCRpMHC system ., In our dataset we found that while all the peptides are at least weak MHC binders the more immunogenic ones tend to have a marginally lower binding affinity to MHC than less immunogenic ones ( Figure 2A ) ., At first glance this seems to contradict the notion 3 that peptide/MHC binding is good indicator for a potential T cell epitope ., However , even if their binding is marginally weaker , those APLs are still at least medium binders for HLA-B*08:01 and this has been shown to be sufficient to be immunogenic 62 ., This tendency of more immunogenic peptides to be slightly weaker binders might be due to the previously described tendency of immunogenic peptides to have larger and more aromatic residues in the central and non-anchor positions 12 ., While these residues enhance the interaction with the TCR , they may reduce , but not impair , the MHC binding affinity by increasing the entropy of the bound state ., To test whether this finding can be generalized we performed an analysis of all available experimental peptide/MHC and T cell activation data from the IEDB 63 ., If all experimental matches ( MHC binding affinity data and T cell activation data are available for the same peptide/MHC combination ) are taken into account the correlation coefficient between T cell activation and peptide/MHC binding affinity is weakly positive ( rPearson\u200a=\u200a0 . 23 , rSpearman\u200a=\u200a0 . 18 ) ., However , if only those matches are taken into account where the peptide is known to bind to the MHC ( IC50<500 ) then the correlation drops to a slightly negative value ( rPearson\u200a=\u200a−0 . 09 , rSpearman\u200a=\u200a−0 . 10 ) ., This would be in agreement with our finding that more immunogenic peptides , if they are at least weak binders , have a marginally lower binding affinity to MHC than less immunogenic peptides ., Often avidity between pMHC and TCR is seen as crucial for the induction of an effective immune response 11 ., However , state-of-the-art computational resources are orders of magnitude away from simulating the formation of a whole immunological synapse 64 ., Hence , we provide insight into the interaction of single TCRpMHC formations ., On the basis of such individual interactions over 100 ns we could not find a strong difference in the binding behaviour between groupM and groupL ( Figure 2B , C ) ., Therefore the findings of our study do not support the notion that each individual more immunogenic pMHC per se necessarily binds stronger to TCRs than less immunogenic pMHCs ., The H-bond network is commonly seen as a major player in modulating interaction landscapes ., Therefore , it would not be surprising if this network is significantly altered between groupM and groupL ., A comparison of the H-bond footprint of the simulations ( Figure, 3 ) with the footprint of the static picture of the first frames ( Figure S3 ) highlights the importance of using the 100 ns simulations ., Several H-bonds are not observed in the static picture e . g . those at the N and C-terminal ends of the peptide ., In contrast the first frames overestimate the number of H-bonds in many other residues ., Hereby , our findings agree well with Reboul et al . who observed fluctuating and transient H-bonds in the TCRpMHC interface 33 ., The H-bond footprints recover several key features of the TCRpMHC binding landscape including dominance of CDRs in the pMHC/TCR interaction , the peptide anchor residues , and the immunogenicity hotspot in peptide position seven ., The H-bonds between the peptide and the MHC and TCR are very similar for the more and the less immunogenic peptide sets ., However , this picture changes for the H-bond footprint between MHC and TCR ., It shows a preference for more immunogenic complexes to have a higher number of H-bonds in the first MHC helix ., For the second helix the picture reverses and less immunogenic complexes have a higher number of H-bonds ( Figure 3D ) ., This indicates a slightly different binding mode of the TCR to groupM and groupL ., This agrees with a chemical shift mapping study that found that different TCRs can create different footprints on MHC helices 65 ., Other authors described a relatively conserved CDR/helix interaction codon 66 ., Some authors even hypothesize that the evolutionarily conserved kinks of MHC helices are central signaling motifs 67 ., Our findings that MHC helices have different H-bond footprints with the TCR if more immunogenic peptides are present further supports the importance of MHC helices for immunogenicity of a pMHC complex ., The closeness of a binding interface is an important property of the mode of interaction ., This closeness is often reflected in the solvent accessible surface area of interface components ., For example Stavrakoudis et al . 37 reported an increased solvation of the LC13/FLRGRAYGL/HLA-B*08:01 interface , Madura et al . observed solvation as important for peptide specificity 68 , and Laimou et al . 23 found that 3G of the immunodominant myelin basic protein ( MBP ) peptide presented by I-Au is more solvent exposed as the analogues 4A and 4Y ., From our data ( Figure, 4 ) it can be seen that the solvation of the interface changes over time and that different CDRs exhibit different distributions of their SASA values ., For example , CDR1α and CDR3α show a broader distribution of their SASA values while CDR2α has rather conserved values ., The SASA distributions for the peptides of groupM and groupL are both slightly positively skewed and almost normally distributed ., Furthermore the first MHC helix has on average a considerably lower solvent accessible surface area than the second helix ., However , while there are strong differences in the solvation of the individual parts of the TCRpMHC interface , we found no evidence that an increased/decreased solvation of any part of the interface takes place for groupM in contrast to groupL ., This is surprising because if TCR/pMHC binding affinity is an important determinant of peptide immunogenicity one might expect the interface solvation to be affected ., For example , MMPBSA 69 free energy calculation methods for the TCR/pMHC interface 70 contain solvation as an important term ., Often authors of MD papers hypothesize that an increased or decreased flexibility of certain parts of the TCRpMHC interface is the underlying principle discriminating immunogenic from non-immunogenic TCRpMHC interactions ., For example Narzi et al . observed an increased flexibility of the ankylosing spondylitis-associated HLA-B*27:05 in contrast to the non-associated HLA-B*27:09 20 ., Furthermore they found that the entropy of a viral peptide was increased compared to self peptides ., Kumar e

Introduction, Methods, Results, Discussion

The interplay between T cell receptors ( TCRs ) and peptides bound by major histocompatibility complexes ( MHCs ) is one of the most important interactions in the adaptive immune system ., Several previous studies have computationally investigated their structural dynamics ., On the basis of these simulations several structural and dynamical properties have been proposed as effectors of the immunogenicity ., Here we present the results of a large scale Molecular Dynamics simulation study consisting of 100 ns simulations of 172 different complexes ., These complexes consisted of all possible point mutations of the Epstein Barr Virus peptide FLRGRAYGL bound by HLA-B*08:01 and presented to the LC13 TCR ., We compare the results of these 172 structural simulations with experimental immunogenicity data ., We found that simulations with more immunogenic peptides and those with less immunogenic peptides are in fact highly similar and on average only minor differences in the hydrogen binding footprints , interface distances , and the relative orientation between the TCR chains are present ., Thus our large scale data analysis shows that many previously suggested dynamical and structural properties of the TCR/peptide/MHC interface are unlikely to be conserved causal factors for peptide immunogenicity .

Immune cells in the human body screen other cells for possible infections ., The binding of T-cell receptors ( TCR ) and parts of pathogens bound by major histocompatibility complexes ( MHC ) is one of the activation mechanisms of the immune system ., There have been many hypotheses as to when such binding will activate the immune system ., In this study we performed the , to our knowledge , largest set of Molecular Dynamics simulations of TCR-MHC complexes ., We performed 172 simulations each of 100 ns in length ., By performing a large number of simulations we obtain insight about which structural features are frequently present in immune system activating and non-activating TCR-MHC complexes ., We show that many previously suggested structural features are unlikely to be causal for the activation of the human immune system .

computer and information sciences, computational chemistry, molecular dynamics, antigen processing and recognition, biology and life sciences, immunology, physical sciences, chemistry, computerized simulations